Research On Computing Offloading And Resource Allocation Strategy In Edge Computing

With the rapid development of 5G and the advent of the Internet of Things era,more and more programs are running on terminal devices,and there are a large number of computing-intensive and bandwidth-intensive tasks.However,terminal devices are limited in computing resources and energy consumption,and completing tasks often requires high latency and high energy consumption.Traditional centralized cloud computing has been unable to provide low-latency,high-efficiency,and low-energyconsumption services for terminal devices.In order to meet the needs of end users,edge computing came into being.Edge computing utilizes computing resources and storage resources deployed in network nodes to provide near-end services for terminal devices,thereby reducing task transmission delay and improving service quality.However,the computing and storage resources of edge nodes are also limited.Reasonable resource allocation can reduce the processing delay of terminal tasks and ensure the quality of services.Thesis studies the computing offloading and resource allocation problems in the collaborative edge computing scenario based on "cloud-edge-device",and the problem based on collaborative caching.It mainly includes the following two research contents:First,thesis studies the task offloading strategy in edge computing based on the resource constraints in the edge computing network.Aiming at the problems of excessive task delay and excessive energy consumption of terminal equipment in the current edge computing offloading strategy,a proposed method based on " The task offloading and resource allocation optimization strategy of cloud-side-device collaboration.Considering the limited computing resources of edge servers and the limited bandwidth of small base stations in the service area,thesis establishes a non-convex optimization model aiming at minimizing the task completion delay and the cost of terminal energy consumption.An optimization algorithm based on simulated annealing is proposed and solved.The simulation results show that the proposed strategy can effectively reduce the system cost compared with the comparison scheme,which verifies its effectiveness.Second,thesis studies the edge cache strategy based on the limited cache space of edge cache nodes.Aiming at the mismatch between the cached content of edge servers in the service area and the preferences of end users in the service area,an edge cooperative caching strategy based on the combination of end user preferences and the popularity of the whole network is designed.The edge cache server in each service area selects the cached content according to the long-term and short-term preferences of end users in the service area and the popularity of the content on the entire network.Considering the limited storage space of the edge server,thesis establishes a non-convex optimization model aiming at minimizing the overall delay of the system,and presents and solves the particle swarm optimization algorithm based on adaptive simulated annealing.Finally,the effectiveness of the proposed algorithm in terms of system delay and cache hit rate is verified by simulation experiments.