Research And Implementation Of The Data Transfer Framework In The Environment Of Mobile Cloud Computing

Along with the rapid development of hardware and software of smartphones which as the representative of mobile devices, the communication and processing capabilities of smartphones are much more powerful, the smartphone has been gradually treated as an essential tool for information processing more than a basic communication tool. However, compared with the PC the mobile device is still resource-constrained, especially in battery capacity. At present, accessing the Internet through wireless network becomes the main source of electricity consumption of smartphones, the limited battery capacity and user applications for the growing demand for power consumption make the energy problem of smartphones more and more imperative. How to make the process of large amounts of data transfer saving power, extending the battery life of smartphones, ensuring the user's daily use is an urgent need to address the problem. Based on this, the research of this thesis focuses on how to make the process of data transmission more energy-saving.On the issues above, this thesis researches and implements the data transfer framework in the environment of mobile cloud computing, using the superiority of the cloud in computing and storage resources and the characteristic of grasping the global trend, combining the local context information of mobile devices when transferring a large amount of data to save energy from the process of transmission. The researches of the thesis are as follows:(1) How to choose a wireless network connection on the mobile device side. There are a variety of wireless network interface for data transmission on mobile devices (EDGE, 3G, WiFi), the availability, data transmission rate and energy consumption of these varieties of wireless networks are quite different. Based on the energy cost model of data transmission, we propose a link selection algorithm, using delaying the data transfer to wait for less energy consumed WiFi network so as to achieve the purpose of saving power.(2) How to adapt to the diversity of data transmission on the mobile device side. The diversity of transmission is caused by real-time data transfer requirements of users, Based on priority, we take different transfer strategies for the different demand of data transmissions. For real-time transmission demand we use the transferring the data immediately, and for the non real-time transmission demand we use the method of delay transmission.(3) How to make the auxiliary work to support the process of data transmission on the cloud side. On the one hand, we use the advantages of cloud on computing and storage resources, partition the task statically, moving the complex part of computing task to the cloud, extending resource-constrained mobile devices. On the other hand, using the characteristics of grasping the overall situation and historical information of the cloud, we propose a delay decision-making algorithm with cloud auxiliary, adopting the Markov prediction model to predict the opportunity of future low-power transmission based on the historical information of users, assisting the delay decision-making of transmission on mobile devices.Based on the content of these studies we design and implement data transfer frameworks in the environment of mobile cloud computing, and deploy the framework application on smartphones and cloud server. The framework application is verified by the actual availability and performance. Experimental results show that using the data transfer framework can save power effectually.