Research On Task Offloading Strategy Of Mobile Edge Computing Based On UAV

With the iterative update of mobile communication technology,Internet of Things(Io T)applications are gradually diversified and personalized,and people's daily life is more and more closely connected with Io T devices.However,Io T devices are affected by physical size,and the computing resources and stored energy are limited,which brings challenges to the development of Io T.The timely emergence of mobile edge computing allows Io T users to offload computing tasks to the network edge computing nodes to run as an effective solution to reduce latency and reduce energy consumption.However,the ground base stations(BSs)with fixed locations and high deployment costs limit the development of mobile edge computing.Meanwhile,the combination of unmanned aerial vehicle(UAV)with strong mobility and low cost and mobile edge computing has important research significance and application value.Aiming at the challenges existing in the existing mobile edge computing,such as over-reliance on ground BSs,UAV-related constraints,and ground device mobility,this paper carries out the following two aspects of research:(1)To address the problem that traditional mobile edge computing is too dependent on ground BSs,this paper proposes a task offloading strategy based on single UAV-assisted mobile edge computing.First,in the single UAV-assisted mobile edge computing network,a joint task offloading decision and UAV location optimization problem is constructed with the objective of minimizing the total energy consumption of ground user devices,considering the limited computational delay,computational resources and UAV energy.Meanwhile,this paper proposes a task offloading strategy based on the alternating direction method of multiplies(ADMM)algorithm.Numerical results show that the proposed task offloading strategy based on ADMM algorithm can effectively reduce the total energy consumption of ground user devices.(2)Since the application scenario of single UAV-assisted mobile edge computing in the first part of the work is limited,this paper further investigates the task offloading strategy based on multi-UAV-assisted mobile edge computing.In this paper,the ground BS in the first part of the network model is discarded and the single UAV is expanded into multiple UAVs,while the mobile characteristics of ground devices are introduced.Under the constraints of time delay and UAV-related constraints,the problem is constructed as a joint optimization problem of task offloading decision and UAV position to achieve the goal of minimizing the total energy consumption of ground mobile devices.Meanwhile,this paper proposes a task offloading strategy based on the asynchronous advantage actor-critic(A3C)algorithm.Numerical results show that the task offloading strategy proposed in this paper has more advantages than the traditional algorithm,and can effectively reduce the total energy consumption of ground mobile devices.