UAVs Employment And Task Scheduling Based On Mobile Edge Computing

Based on the ideas and concepts of the Internet of Things and cloud services,Edge Computing(EC)was proposed as a model to process and analyze data through network edge nodes.It is widely used in specific fields such as cloud offloading,face recognition,smart home,smart city.As a part of EC,Mobile Edge Computing(MEC)takes the advantages low latency and high bandwidth of edge computing in making deployment strategies and offloading decisions,which has become a significant study of edge computing application research at home and abroad.However,effected by different application scenarios,numerous data and various node selections,MEC causes problems such as massive resources waste resulted from excessive network delays and high bandwidth in the application.Therefore,the core issue of mobile edge computing lies in the reasonable deployment of edge cloud and the task schedule of mobile users.If the core issue of edge cloud is well settled,mobile edge computing can build a service environment with high performance,low latency,and high bandwidth to reduce system energy consumption as much as possible.By developing and optimizing the existed algorithms of large-scale mobile node allocation and scheduling,this paper constructs a multi-UAV deployment and task scheduling system based on MEC.The main innovations of this system are as follows: First,a double-layer optimization method of EMC and a new quadruple encoding mechanism are proposed,so the UAV deployment and mobile user task scheduling are comprehensively optimized to reduce the overall energy consumption of the system.Secondly,an adaptive bee colony algorithm is proposed.elimination operator factor and difference strategy library are employed in the upper-level optimization algorithm and the difference vector is used to enhance the search ability of the algorithm,and balance the local optimization-searching and global search so that mobile nodes can autonomously select new strategies and the resource allocation can be optimized.This algorithm is applied to the deployment optimization of multiple UAVs.Thirdly,an efficient greedy algorithm is proposed.In the lower-level optimization,the task scheduling of multiple mobile nodes is transformed into a 0-1 integer programming problem,which can classify tasks according to the scale of the system to obtain an approximate optimized solution in less time.By implementing a simulation experiment of large-scale UAVs,this paper finds that the EMC two-layer optimization method and the algorithm of large-scale mobile user deployment and task scheduling system proposed in this paper improve the success rate of tasks,reduce the number of required drones,and reduce the comprehensive energy consumption of system.