移动边缘计算中的端到端任务分配算法

为了提高移动应用程序的运行效率，移动边缘计算将部分任务从终端设备迁移到边缘云中计算来缩减应用程序的运行时间和终端设备的能耗。针对应用程序所需的总代价即能耗和时间两个目标进行了研究，提出一个移动边缘计算模型和基于贪心策略的快速算法（HGA）；构造了一个结合贪心策略的粒子群（HPSO）算法，进一步优化HGA的解。实验结果表明，与传统所有任务只在一个设备上执行和尽可能上传云端执行两种策略相比，提出的HGA总代价分别优化28.5%和9.1%；与HGA相比，HPSO算法总代价减少12.3%；即所提算法能有效减少系统的总代价，更加满足用户需求。

Algorithm for D2D task allocation in mobile edge computing

Tasks of the mobile devices are offloaded on the edge servers to ensure the efficiency of the mobile applications in mobile edge computing. This research aimed to optimize the energy and time consumption of the mobile device by proposing a computing model and two heuristic algorithms. One is fast greedy algorithm(HGA); the other is particle swarm optimization algorithm based on HGA(HPSO) in which the solution of HGA was further optimized. Experimental results show that the solution quality of HGA is reduced 28.5% and 9.1% in terms of the total energy and time consumption with the strategies of all tasks on one device or as much as possible on cloud servers. The solutions quality of HPSO is improved up to 12.3% in comparison to HGA. The proposed algorithms can effectively reduce the total cost of the system and meet the needs of users.