Research On Task Offloading And Resource Allocation Of Multi-node Cooperation In Edge Computing

With the rapid upgrade of wireless communication technology,the popularity of smart terminals,the rapid growth of mobile data,and the rise of diversified services,higher demands are proposed on battery capacity,computing and storage resources.The traditional cloud computing model cannot meet the real-time requirements for computing-intensive and delay-sensitive applications.As a key technology of the fifthgeneration communication network(5G),mobile edge computing(MEC)first uploads terminals' computing tasks to adjacent edge servers,then MEC servers compute and transmit the results back to the terminals.Compared with the cloud center,MEC system greatly decreases latency and improves user experience.Although MEC can greatly reduce the delay,energy consumption and the system overhead,the computing and storage resources of the edge server are limited.Therefore,how to efficiently offload tasks to suitable edge nodes and allocate optimal computing and wireless resources for offloading tasks has important research significance.Taking into account the importance of task offloading and resource allocation in MEC,the thesis conducts indepth research on task allocation and collaborative optimization of multiple resources in multi-node edge computing networks.And take system overhead and system processing capacity as performance indicators to improve system performance while meeting user service quality.The main research contents are as follows:(1)Aiming at the problems of high system overhead and obvious delay jitter when intensive tasks are offloaded in the MEC system,a strategy of fine-grained task division and multi-edge node resource allocation is proposed.This strategy first analyzes the system task execution overhead and the resource allocation mechanism of the end user under the constraints of system delay.Secondly,a joint convex optimization goal based on computation offloading and task allocation is established.And finally,the Lagrangian multiplier method is used to iteratively update to obtain the optimal solution.The simulation results show that the proposed task offloading and resource allocation scheme reduces the task execution overhead while ensuring user service quality,and effectively improves the MEC system performance.(2)In response to the problem of low spectrum efficiency and resource utilization caused by mass access of 5G mobile networks,non-orthogonal multiple access technology is introduced into the MEC system.By studying the problem of multi-user,multi-task and multi-server computing offloading,we jointly optimize task offloading and resource allocation to improve the processing capacity of the system.The proposed problem is described as a mixed integer nonlinear programming problem.First,the objective optimization problem is decomposed into two sub-problems of resource allocation and task allocation.Secondly,the problem of resource allocation is further decomposed into computing resource optimization and communication resource allocation.For the allocation of communication resources,a fixed power allocation is performed first of all,and then the sub-channel allocation problem is regarded as a many-to-one matching problem between sub-channels and users.In addition,a lowcomplexity sub-optimal matching algorithm for sub-channel allocation is proposed to maximize storage efficiency.Based on the proposed sub-channel allocation scheme,the transmission power allocation is regarded as a convex optimization problem,which can be solved by the Lagrangian multiplier method.Finally,under the condition of resource allocation,all terminal equipment tasks are allocated.Experimental results show that the proposed scheme can effectively reduce network delay and energy consumption,improve system processing capacity,and further improve MEC system performance.