Research On UAV Edge Computing Task Scheduling Algorithm Based On Intelligent Optimizatio

The development of the Internet of Things makes data services spread to every corner of the human society,driving social progress but also bringing huge data increment and communication service demand,which poses a severe challenge to wireless communication technology.The UAV-assisted mobile edge computing technology provides computing services for terminal equipment at the network edge.It inherits the advantages of short communication delay,good confidentiality and good scalability of edge computing,and at the same time has the advantages of high mobility,flexible deployment and low cost,which can effectively alleviate the problems of long-distance data transmission and mass data offloading.However,for the UAV-assisted mobile edge computing technology,how to achieve the purpose of low delay and low energy consumption through task scheduling and UAV planning is also a problem that needs to be considered.Therefore,this thesis focuses on the task scheduling problem of mobile edge computing system,and proposes a task scheduling scheme with low delay and low energy consumption.The research content and main achievements of this thesis are as follows:(1)This thesis studies the task offloading problem in multi-UAV cooperative mobile edge computing system,and proposes a mobile edge computing offloading calculation scheme which takes offloading decision and UAV position deployment as optimization variables.To minimize system delay,the task offloading optimization problem is constructed.Distributed deep neural network was used to solve the minimum delay offloading decision scheme,and then the offloading scheme based on the neural network was solved to solve the minimum delay UAV deployment scheme of the system.The cycle was repeated until the algorithm convergence.Simulation results show that the proposed optimization algorithm can effectively reduce the processing delay of the system.(2)The task offloading strategy of UAV-assisted mobile edge computing based on digital twin is studied,and a mobile edge computing offloading scheme is proposed to optimize the offloading decision,equipment transmission power and UAV computing resources.With the goal of minimizing the processing power of tasks in the system,the task offloading calculation optimization problem was constructed.The deep reinforcement learning was used to calculate the offloading strategy of the equipment.To optimize the equipment transmission power and computing resource allocation of UAV based on offloading scheme which was explored by deep reinforcement learning,the genetic algorithm was utilized.Simulation results show that the proposed optimization algorithm can effectively reduce the system energy consumption.