Research On Task Scheduling Strategy For WPT Multi-access Edge Computin

Muti-access Edge Computing migrates computing and storage resources to the edge of the network near the Internet of Things devices,effectively avoiding high latency in data transmission between devices,reducing data transmission costs,and improving network bandwidth utilization.In order to solve the problem of limited battery reserves and computing resources of Internet of Things devices,Wireless Power Transfer technology is applied to MEC scenarios,which can improve the resource utilization efficiency of Internet of Things devices and achieve sustainable and efficient computing.In WPT-MEC wireless networks,designing efficient and reasonable task scheduling policies is the key to improve task execution efficiency,reduce system computing load and improve network performance.Firstly,aiming at the task scheduling problem of single edge node in WPT-MEC network,the randomness of task and energy arrival and the uneven distribution of computing resources of edge server were analyzed,and a dynamic task scheduling model with queued delay constraint was established.By using Lyapunov optimization method,the dynamic random optimization problem was transformed into static optimization subproblem in independent time slot,and Efficient Task Scheduling and Resource Allocation Algorithm was proposed to obtain a low energy task scheduling policy.Simulation results show that the proposed ETSRA algorithm can reduce the system energy consumption by approximately 49% while maintaining the stability of the system task queue.Secondly,considering the coupling between task scheduling and wireless energy transmission,a collaborative optimization model of local computing resources and transmission delay was established for the task scheduling problem of UAs in WPTMEC network.A Low Energy Consumption UAV Cooperative Task Scheduling Algorithm was proposed by using Lyapunov optimization method,Lagrange duality method and Newton interior point method.To obtain the task scheduling strategy that minimizes the system energy consumption.Simulation results show that the proposed LEUCT algorithm can effectively reduce the system energy consumption by approximately 27% and reduce the task queue redundancy by approximately 19%.Finally,aiming at the task scheduling problem of the hollow-earth integration in the WPT-MEC network,the influence of the dynamic behavior of the Internet-ofThings device association matching UAV and the difference of the deployment of UAV computing resources on the task scheduling strategy was comprehensively analyzed,and a three-layer network architecture model of the collaboration between the Internetof-Things device,UAV and low-orbit satellite was established.Using Lyapunov optimization method and many-to-one matching game,the Dynamic Optimal Association-matching Task Scheduling Algorithm is proposed.To obtain a task scheduling strategy that minimizes the computing cost of system resources.Simulation results show that the proposed DOATS algorithm can effectively reduce the system cost by approximately 34%.