Research On Offloading And Allocation Of Computing Resources In Mobile Communication

At present,with the continuous development of the intelligence of mobile communication,the computing task in mobile communication is more extensive.Therefore,it is necessary to make effective management and rational utilization of this computing resource in mobile communication.The computing tasks in mobile devices are more and more along with the improvement of mobile devices,and the problem of insufficient energy of mobile devices becomes more obvious.Besides,the soft base station based on general purpose processor(GPP-based SBS),which is a flexible high-compatible programmable base station,is more depended on the computing power of general processor.Therefore,it is need to make rational utilization of general purpose processor.Based on the above two aspects,this thesis investigates the energy-optimal offloading of mobile computing and computing resource allocation in GPP-based SBS.The main research contents are summarized as follows:For the situation that the new wireless access network can better support the mobile computing offloading,this thesis analyzes the typical software structure of the mobile computing offloading and gives the model of energy-optimal mobile computing offloading.Then,the advantages and challenges of mobile computing offloading based on heterogeneous networks are analyzed.Aimed at the problem of offloading strategy,the energy consumption in offloading and the feasibility of saving energy of mobile devices through computing offloading are analyzed.Finally,the model of energy-optimal mobile computing offloading is presented.Considering that general optimal solution of offloading has high complexity,a mobile computing offloading algorithm based on State Transition is proposed.This algorithm is based on the approximate solution of offloading based on Greedy Strategy.Then the Bounding algorithm is used to reduce the problem dimension and the search space.And the remaining problem is solved by the offloading algorithm based on Dynamic Programming finally.The simulation results show that the optimal solution can save energy of mobile device around 40%.For the inadequate general computing resource utilization in GPP-based SBS,the computing resource allocation in GPP-based SBS for high density services is designed.First the mapping and partition of computing resource is proposed.Then,the model and algorithm of computing resource allocation of GPP-based SBS are given.The simulation results show that the computing resource algorithm can improve computing resource utilization about 10%.In conclusion,this thesis investigates the offloading and allocation of computing resource in mobile communication from the above three aspects.The offloading algorithm and allocation algorithm are proposed on the basis of existing research and the algorithms are verified through simulations.