Research On Performance Optimization Of UAV-based Mobile Edge Computing System

In this era of interconnection of all things,drones have become an indispensable part of the Internet of Things due to their flexible deployment and strong mobility.With the advent of the 5G era,mobile users have continuously increased their requirements for network transmission speed and stability.In certain areas where the number of users has increased sharply,such as large-scale concerts and sports events,traditional ground base stations are often overwhelmed,and increasing the deployment of base stations will greatly increase the cost of operators and generate unnecessary expenses.At this time,the use of drones equipped with micro-cell base stations to fly over the area to assist ground communication base stations to provide network services to users can reduce the cost of infrastructure construction on the basis of solving the problem.In addition,edge computing technology has also attracted great attention from researchers in industry and academia in recent years.Different from cloud computing,edge computing deploys a part of computing nodes at the edge of the network to shorten the data transmission distance and thereby reduce latency.However,edge computing servers are usually fixed on some specific nodes,which does not substantially help users who are far away from these nodes.Therefore,using drones to fly over a specific area with edge computing servers and provide computing offloading services for ground users has become an emerging way to solve this problem.This paper mainly considers how to improve the computational capacity of the entire system under multiple constraints such as limited energy resources and communication resources in a mobile edge computing system based on UAVs.In addition,the optimization problem of UAV energy consumption in this system is analyzed and researched.The specific work is as follows:1.In the scenario of drones flying with edge computing servers flying over specific areas to provide computing offloading services for ground users,the drone’s flight trajectory,bandwidth resource allocation,and offloading time and local computing time allocation are combined.Optimization,a reasonable optimization algorithm is proposed and continuous convex approximation and other techniques are used to maximize the total calculation amount of the system.Finally,the results of simulation experiments show that the total calculation amount of the system is the largest compared with other benchmark schemes.Especially when compared with the fixed local calculation and task offloading time scheme,the total calculation amount of the system is increased by nearly 100%.2.In the mobile edge computing system based on drones,the problem of minimizing the energy consumption of drones while providing computing offloading services according to the different Qo S requirements of ground users is studied.Under the optimal scheme,the optimal flight trajectory of the UAV and the optimal allocation strategy of resources are given.It can be seen from the simulation results that the optimization scheme proposed in this paper is more effective in reducing the energy consumption of the UAV compared with the benchmark schemes such as fixed UAV flight trajectory.