Research On Resource Allocation And Power Control In Mobile Cloud Computing

With the rapid development of the mobile Internet technology and hardware, cloud computing, as the most popular direction of computer science, ushered the combination with the mobile internet. Mobile cloud computing, which benefits from the advantage of mobile network and cloud computing, also has to face those problems like signal question of the mobile network and energy consumption of the mobile equipment. As a new mobile cloud computing architecture, cloudlet was designed with the purpose of solving above problems by add a middleware between the cloud service provider and mobile user.Based on the analysis of cloudlet in the mobile cloud computing, this paper carries out the investigation and analysis on mobile cloud computing, cloudlet, virtual resource management technology, and mobile network power control research. Cloudlet, as a cloud computing platform with limited resources, faces the fundamental problem of the confliction between maximize the resource utilization and meet the demand of users. At the same time, according to the scale of users' transmit power, user gets different transmit rate and energy consumption, this leads to a non-cooperative game between users who wants to minimize their energy consumption. By analyze the allocation of resources and power control problem on the mobile cloud computing which based on the cloudlet, the final problem is transformed into an optimization question includes resource allocation, admission control and user power control game.After analyzing the common optimization ideas and the complexity of our problem, we decide to use the intelligent optimization algorithm to deal with the problem this paper present. In comparing the advantages and disadvantages of several intelligent algorithms, considering the fitness to our problem, we chose the ant colony algorithm. In order to meet the goal of the optimization problem in the mobile cloud computing, we put forward several improvement on the traditional ant colony algorithm. We carry on several experimental, which results show that, the algorithm has a good performance in both optimization results and convergence speed. Finally, through the simulation experiment on the change of some parameters in the algorithm, we give some solutions on how to improve the algorithm performance in this problem.