A Research Of Resources Allocation Algorithms For The UAV-based Mobile Edge Computing

Compared with cloud computing,mobile edge computing technology reduces communication delay and improves the user experience.As a kind of low-cost aircraft,UAV can be used as a mobile communication platform to provide effective communication coverage for remote areas.The UAV-based Mobile edge computing technology,which can extend the service coverage and save the cost of base station deployment,has become a leading research direction.At the same time,the Internet of things devices are inconvenient to be recharged in the remote area.The UAV can use wireless charging technology and becomes a moving platform to provide energy supply.Although the UAV can fly close enough to users,the MEC service is limited by communication,computing and energy resources.How to allocate its onboard resources effectively is the key to this research.Based on a vertical research project,the emphases of this paper are as follows:1)This paper analyzes the characteristics of the UAV-based mobile edge computing technology.It studies the UAV-based mobile edge computing network and determines the role of the UAV in the mobile edge computing system.2)In order to minimize the energy consumption of a single UAV,the hovering design based joint energy resource allocation algorithm is proposed.3)To deal with the co-existence of multi-tasks in the UAV-based mobile edge computing system,a multiple workflows based structure is proposed.Inspired by the flow shop problem,a joint resource allocation algorithm is proposed.4)For the multiple UAVs based MEC system,a heuristic path planning algorithm is proposed.Moreover,in order to minimize energy consumption,a joint resource allocation algorithm is proposed.The simulation results show that the joint resource allocation algorithm is effective.Compared with the benchmarks,the proposed joint resource allocation algorithm has better performance on system energy efficiency and can extend the flight time of the UAV.Compared with the traditional greedy algorithm,the path planning based energy allocation algorithm can obtain better solutions.The proposed algorithm has a faster convergence speed compared with the genetic algorithm.