Unmanned aerial vehicle assisted mobile edge calculation method, device, and control equipment

Abstract

The invention discloses an unmanned aerial vehicle assisted mobile edge calculation method, a device and control equipment. The method comprises the steps of initializing user transmission power, unmanned aerial vehicle CPU frequency and unmanned aerial vehicle position; collecting an optimal user unloading decision under the condition; the optimal user transmission power, the unmanned aerial vehicle CPU frequency and the unmanned aerial vehicle position under the current optimal unloading decision are obtained, and a current system utility function is calculated; based on the obtained user unloading decision, transmission power, energy consumption and time delay constraint, unmanned aerial vehicle calculation resource constraint and position constraint, obtaining user transmission power, unmanned aerial vehicle CPU frequency, unmanned aerial vehicle position and unloading decision when the utility function is maximum; and the unmanned aerial vehicle is controlled to fly to the optimal position and is allocated with the optimal calculation resource, and the unloading user is controlled to transmit with the optimal transmission power. By adopting the method provided by the invention to control the user and the unmanned aerial vehicle, the energy consumption of the user and the calculation task completion time can be reduced, and the calculation efficiency of the user is improved.

Description

technical field [0001] The present invention relates to the technical field of drones, in particular to a method, device and control equipment for mobile edge computing assisted by drones. Background technique [0002] In order to meet the high-bandwidth and low-latency requirements required by the rapid development of mobile networks and reduce network loads, Mobile Edge Computing (MEC) has become one of the key technologies for 5G / 6G networks. [0003] Due to its stable operation, flexible movement, and strong environmental adaptability, UAVs are increasingly considered in the field of edge computing to assist ground equipment in computing. [0004] In the existing technology, when the ground user cannot complete the computing task in real time (such as post-disaster, hotspot scene), the ground user can choose to offload the computing task to the UAV side for calculation, and the UAV will send back after completing the computing task. [0005] However, considering that th...

AI Technical Summary

Problems solved by technology

[0008] The slow speed is mainly reflected in two parts. One is that the algorithm to complete the optimal decision is complex and takes a long time; the other is that it takes a long time to complete all computing tasks.

In the existing technology, the enumeration method or the independent optimization of various indicators for a single user are considered, and the former has high algorithm complexity (O(2 K )), for a large number of users, the calculation time is very long, and the latter needs to judge each user one by one. In the case of limited remote UAV computing resources, it takes a long time for a single user to complete the task, resulting in the overall completion of the task. long time

[0009] 2. Low system efficiency

[0010] Low efficiency is mainly reflected in two parts, one is the low energy efficiency of users; the other is that the system utility function is smaller

Existing technologies consider performing offloading tasks locally or independently offloading each user. Since offloading can save energy consumption at the user end, local execution consumes the most energy, while independently offloading each user consumes the least energy. However, due to The global resources of the system are not considered, so the energy efficiency is low

In terms of system utility function, since the existing technology allows users to choose to offload to remote computing tasks as much as possible, it may cause the system utility function to be negative (that is, the calculation time is too long, the offloading task fails), and even worse than choosing the local Worse case of execution

[0011] 3. Low robustness of the system

[0012] In the case of limited remote computing resources and a large number of ground users, the existing technology considers the optimization of a single user and ignores the problem of limited resources. There may be a situation where a large area of ​​users cannot complete computing tasks in time, and the system stability is relatively low. Difference

Images