Research On Mobile Cloud Service Mechanism And Computation Offloading For 5G

With the advent of the era of 5G "Internet of Everything",cloud computing platform,mobile wireless communication network and Internet together constitute the core architecture of mobile cloud computing,the technology will make computers resource sharing in a wireless environment.At present,the popularity of smart mobile terminals and the increasing demand for computing resources in mobile applications,the limited problem of mobile terminal resources(including computing,storage,energy,etc.)has become prominent.Computation offloading achieves the goal of resolving and expanding resources by migrating part of intensive tasks from local to the cloud computing nodes.Based on the characteristics of 5G network architecture,this paper proposes a distributed mobile cloud computing collaborative architecture,analyzes the collaborative requirements,and introduces collaborative controller components.Through analysis of 5G mobile network technology,The foreseeable core technology provides the feasibility support for the mobile cloud computing architecture,embedding the mobile cloud resources into network elements such as radio access network base stations,providing near-end low-power and low-latency cloud services;introducing a distributed mobile cloud computing collaborative controller.In order to meet the current multi-site task migration requirements,we builds the computational node model and delivery model of the edge cloud based on the open system interconnection migration model.Through the analysis of application requirements,the computing power of the mobile terminal and the channel state,the transmission power of the mobile terminal and the occupation of the CPU computing resources by the application are optimized.Combined with computing task information,cloud information and network channel information,a multi-site computing migration method is designed.Based on the analysis of Markov decision process,the observable Markov decision process is further studied.Finally,the mathematical model is used to build the simulation scenario,and the performance of the algorithm is analyzed based on the obtained performance parameter data.