无人机辅助的服务缓存边缘计算最优计算卸载决策与资源分配

UAV-enabled Service Caching Edge Computing Optimal Computation Floading and Resource Allocation Strategy

边缘计算中利用无人机作为边缘节点进行动态部署,能够适应复杂的环境,大大提升边缘计算系统的性能.本文提出利用无人机辅助的服务缓存边缘计算最优计算卸载和资源分配策略.此策略在确定无人机3D位置和边缘服务器中服务的部署,以实现在时延约束下最小化能耗的目的.具体来说,首先,建立本地计算模型和MEC计算模型,计算出任务的处理时延和能耗;其次,在服务缓存、时延约束等条件下,建立最小化能耗的数学模型;最后,采用遗传算法框架对目标问题进行求解.求解过程采用双层优化方法,外层层将无人机3D位置和服务缓存方案放入基因编码,内层先利用贪心的思想确定资源分配,再将问题转化为整数线性规划问题进行求解.通过仿真证明了本文所提出算法的可行性和优越性.

In edge computing,unmanned aerial vehicles(UAVs)are used as edge nodes for dynamic deployment,which can adapt to complex environments and greatly improve the performance of edge computing systems.This article presents an optimal computing offloading and resource allocation strategy for edge computing using UAV-enabled service caching.This strategy is to determine the 3D location of UAV and the deployment of services in edge servers to achieve the goal of minimizing energy consumption under delay constraints.Specifically,first,establish a local computation model and an MEC computation model to calculate the processing delay and energy consumption of the task.Secondly,establish a mathematical model to minimize energy consumption under the conditions of service caching and delay constraints.Finally,The genetic algorithm framework is used to solve the target problem.The solution process adopts a two-layer optimization method.The outer layer puts the UAV′s 3D location and service cache scheme into the genetic code,and the inner layer first uses the greedy idea to determine the resource allocation,and then converts the problem into an integer linear programming problem for solution.The simulation proves the feasibility and superiority of our proposed algorithm.