Energy-optimal Application Offloading Scheme In Mobile Cloud Computing

With the rapid development of mobile communications,mobile devices have become indispensable parts of people's daily life.Due to the limited physical size,however,mobile devices themselves cannot fully meet the requirements of applications on computation,storage,battery life and etc.Mobile cloud computing(MCC)technology is an Internet-based computing mode,which allows users to rent real-time computing services of IT resources such as servers,software and networks remotely through the Internet.With mobile cloud computing,mobile devices can upload applications to the cloud and authorize it for processing.This process is called application offloading.However,compared with the rapid development of computing,battery storage technology has developed slowly.Battery life has become a bottleneck in the development of mobile devices.Therefore,how to optimize the energy consumption of mobile devices during the application offloading is crucial.This thesis focuses on the energy-optimal application offloading scheme in mobile cloud computing.The main contents of this thesis are as follows:1.Firstly,a modified Lyapunov optimization method is proposed.The algorithm can obtain a closed-form expression of the transmission rate only according to the actual measurement of the channel gain in each time slot and thereby can avoid the deviation from prediction of the channel gain in future time slots.The numerical results demonstrate that our proposed algorithm can save about 30% of the energy compared to traditional algorithms.2.Furthermore,due to the rapid development of mobile communications and the variety of services,many different services coexist in the actual cellular cell,sharing the same spectrum resources.Therefore,this thesis also investigates the scenario where multiple mobile devices with MCC and non-MCC services coexist.A bandwidth allocation and the corresponding transmission rate scheduling schemes are proposed with the objectives of simultaneously maximizing the overall system throughput and minimizing the energy consumption of individual mobile device with MCC service.The numerical results show that the proposed algorithm can effectively take into account both the optimization of system throughput and the optimization of energy consumption of each mobile device.