Research On Multi-task Computing Offloading And Resource Allocation Under Heterogeneous Edge Cloud Architecture

With the rapid development of mobile communication technology and the wide application of smart terminals,users have higher requirements for service quality and experience quality.However,the computing power,storage capacity and battery capacity of the existing terminal devices cannot match the emerging applications such as augmented reality and unmanned driving,which are time-sensitive,have a large amount of tasks,and consume a lot of energy.Mobile edge computing provides a solution to the above problems by deploying edge servers with computing,communication,and storage capabilities at the edge of the network.Users can offload tasks that cannot be matched by terminal devices to edge servers for processing.However,computing resources and communication resources are often limited.It is particularly important to allocate resources reasonably and improve the efficiency of resource use.With the goal of minimizing the overhead of the system,this paper mainly studies the multi-task computing offloading and resource allocation under the heterogeneous edge cloud architecture.The details are as follows:The first part of the work of this paper is to run computing-intensive and delay-sensitive applications on terminal devices,build an edge cloud heterogeneous network model,and perform multi-task computing offloading through edge devices such as unmanned aerial vehicles,roadside units,vehicles,and edge cloud servers.In order to reduce the delay and energy consumption,a hybrid optimization algorithm combining particle swarm and genetic algorithm is proposed and applied to the task offloading model of heterogeneous edge cloud.The algorithm improves the selection,crossover,mutation and other operations in the genetic algorithm,and makes up for the defects of the premature convergence and falling into local optimum of particle swarm optimization algorithm through genetic algorithm.The performance of the algorithm is verified through the test of six classic test functions and the comparison experiment with the baseline scheme.The simulation results show that the hybrid optimization algorithm can effectively improve the convergence accuracy and reduce the system overhead.The second part of this paper is to allocate the computing resources and communication resources of edge nodes on the basis of the first part,and to jointly optimize the computing offloading and resource allocation.The computing resources are allocated to each user task through the Lagrange multiplier method,and the communication resources are allocated according to the proportion of offloaded tasks.The calculation offloading and resource allocation are jointly optimized through loop iteration,so as to obtain the best offloading decision and resource allocation scheme.The simulation results show that compared with the independent task offloading and resource allocation optimization,the effect of joint optimization in reducing system overhead is more obvious.