Research On Intelligent Optimization Algorithm Of Network Resources Based On Cloud-Edge Collaboration

With the vigorous development of multimedia services in mobile communications,the surge in wireless network traffic and the emergence of a large number of intelligent terminal devices have resulted in a shortage of network resources.Therefore,the cost of network equipment of Internet Service Provider(ISP)and Content Provider(CP)has surged,and the revenue has decreased.In addition,it is more difficult to provide lowlatency services in latency-sensitive scenarios such as the Internet of Vehicles(Io V),and the service quality cannot be guaranteed.As a solution for future 6G networks,cloud-edge collaboration combines cloud computing with edge computing,while having powerful computing and storage capabilities and short transmission delay,which can greatly meet the diverse needs of users.At the same time,considering the strong heterogeneity of the cloud-edge collaboration architecture and a large number of ubiquitous resources,it is very important to formulate a reasonable resource optimization and allocation strategy.It is an important means to improve system resource utilization,relieve the pressure of equipment expansion,and reduce service delay.Therefore,in order to improve the service benefits of ISP and CP and reduce the Io V service delay,this thesis will study the resource optimization problem under the cloud-edge collaboration architecture.The main research contents are as follows:1.Research on profits optimization under the cloud-edge collaboration architecture based on the Stackelberg game: This paper studies the resource optimization problem of cooperation between ISP and CP in the cloud-side collaboration scenario with the goal of improving the revenue of ISP and CP.On the premise of fully considering the economic behavior and content popularity of ISP and CP under the cloud-edge collaboration architecture,we model the interactive behavior of ISP and CP as a profit maximization problem based on centralized strategy,and optimize cache resources by optimizing cache decisions to maximize network profits.Then,we designed a distributed win-win cooperation scheme based on Stackelberg game and a feasible backward induction method.Finally,the simulation results show that the performance of the proposed Stackelberg game model is equivalent to that of its centralized solution.2.Research on intelligent resource allocation in cloud-edge collaboration assisted Io T scenarios: This paper studies the resource allocation problem in Io V assisted by cloud-edge collaboration for the purpose of further reducing the Io V service delay.We represent the joint resource allocation problem as a delay optimization model based on queuing theory,propose a resource optimization scheme based on deep reinforcement learning.The service nodes make routing decisions and content cache decisions for each request according to the user requests information and the current network resource status.The proposed strategy reduces the network delay and improve the quality service by jointly optimizing the allocation of computing cache and communication resources.Simulation results show that the proposed strategy performs better than existing cloud-edge collaboration solutions and converges quickly in network environments with different parameters.