Research On Edge Computing Offloading Strategy In UAV Emergency Communication Network

With the development of UAV technology,benefiting from its advantages of agility,speed and flexibility,UAV has become a popular flight equipment in various fields,and one of the most important application scenario is emergency communication.In the UAV emergency communication network,the UAV and Mobile Edge Computing(MEC)can be combined to provide task computing offloading services for ground terminals by deploying communication and computing equipment on the UAV.Therefore,the UAV edge computing offloading technology has become a popular research direction at present.The topic of thesis is selected from the national key research and development plan project"Research on Emergency Communication Technology and Key Portable Equipment for Large-scale Regional Major Natural Disasters".In order to reduce the task computing delay and energy consumption,thesis studies the edge computing offloading strategy under the UAV emergency communication network,so as to improve the system performance.In the thesis,two scenarios of air-ground coordination and UAV coordination are considered respectively,and corresponding edge computing offloading strategies are proposed.The main work of thesis is as follows:1)This thesis summarizes the current research status of edge computing offloading in UAV emergency communication network and points out the existing problems.2)Aiming at the UAV emergency communication scenario with diversified task requirements,thesis proposes a UAV edge computing offloading strategy based on air-ground cooperation.Based on the characteristics of emergency communication network,an air-ground cooperation architecture based on ad-hoc network is proposed,According to different task requirements,we define the weighted sum of task delay and energy consumption as system cost.On this basis,the system costminimization problem of task scheduling and multi-UAV deployment is proposed.To solve this problem,we decomposed it to two sub-problems that were solved by proposing a swap matching based task scheduling subalgorithm and a successive convex approximation(SCA)based multi-UAV deployment sub-algorithm.Then two sub-algorithms iterate continuously to obtain the suboptimal solution with low complexity.Finally,the simulation results verify the convergence,feasibility and effectiveness of the proposed algorithm.The simulation results show that compared with the benchmark algorithms,the proposed algorithm can greatly reduce energy consumption and delay,and further prolong the survival time of the UAV network.3)Aiming at the UAV emergency communication scenario with delaysensitive requirements,thesis proposes an edge offloading strategy based on mutil-UAV cooperation.In the UAV edge computing offloading network,a multi-UAV cooperation computing architecture is established,which realizes the cooperation offloading of computing tasks by multi-hop transmission between UAVs.The transmission and computing delay generated during the offloading process of edge computing are considered.The optimization problem of minimizing the total length of time slots is constructed by the way of executing tasks in different time slots.The proposed optimization problem is decomposed into three sub-problems,namely,the computing task scheduling,the computation resource allocation,and the UAV trajectory.To solve these subproblems,thesis proposed the penalty based task scheduling algorithm,the KKT(KarushKuhn-Tucker)condition based computation resource allocation algorithm,and the SCA based multi-UAV trajectory algorithm respectively.Based on iterating the sub-algorithms,a joint optimization algorithm is proposed to minimize the total time slot size.Finally,the simulation results show that,compared with the benchmark algorithms,the strategy proposed in thesis can more effectively utilize the collaborative computing unloading capability among multiple UAVs,and complete the optimization of UAV flight trajectory under the condition of collaborative strategy.Meanwhile,the proposed strategy can significantly reduce the task execution delay and improve the calculation efficiency of UAV mission.