Calculation Uninstall Utility Function Based On Switching Mobile Cloud

Rapid evolution of wireless communication technology brings wireless network bandwidth continues to expand, the current rate of development of wireless access devices is reached astonishing degree, however, the limitations of the mobile terminal compared to a desktop PC computing power, storage, battery power and other aspects still exist, with mobile cloud computing technology, the mobile terminal can be computationally intensive tasks rational allocation of sufficient resources to calculate a proxy server for processing, then the operation is completed the results from the proxy server to retrieve, it can be said that the cloud computing offloading in the wireless network environment is improving terminal operation ability and an important way to reduce energy consumption of mobile terminal.Due to remain the state of mobile wireless mobile terminals under the network environment, the computing tasks of offloading is faced with the problem of network selection and task handover, the thesis take the problem as a starting point, on the basis of the concept of mobile cloud computing offloading, combined with the mobile cloud environment task offloading and research analysis of task handover, task offloading handover strategy is proposed based on the multi-attribute utility, utility analysis theory is applied to the mobile cloud decision making issues, policy embarks from the actual situation, combined with the actual mobile path model and agent network terminal load state analysis of mobile cloud computing environment can be obtained by calculation, and can according to the terminal real-time power state dynamic adjustment offloading handover strategy, realize the dynamic optimization of energy consumption spending.Thesis use Matlab simulation experiment platform, through the experiment achieved four strategies, to verify the feasibility and effectiveness of strategy. Through the analysis of experimental results, the computing offloading handover strategy based on multi-attribute utility can save the cost of offloading task overhead, and when the number of the task and agent network increase, the strategy can effectively reduce the computational tasks offloaded to the agent under the condition of execution time, at the same time can be implemented according to the battery terminal real-time allowance offloading dynamic adjustment of energy consumption calculation.