学术英语论文.docx

1、学术英语论文 NANCHANG UNIVERSITY 课程名称： 学术英语 题 目： A Study of Energy Efficient _ Cloud Computing Powered by_Wireless Energy Transfer _英语班级： 理工1615班 专业/年级： 物联网工程 161班 姓名/学号： (47) 二零一八年六月A Study of Energy Efcient Mobile Cloud Computing Powered by Wireless Energy TransferAbstractAchieving long battery lives or

2、 even self-sustainability has been a long standing challenge for designing mobile devices. This study presents a novel solution that seamlessly integrates two technologies, mobile cloud computing and microwave power transfer (MPT), to enable computation in passive low-complexity devices such as sens

3、ors and wearable computing devices. Specifically, considering a single-user system, a base station (BS) either transfers power to or offloads computation from a mobile to the cloud; the mobile uses harvested energy to compute given data either locally or by offloading. A framework for energy efficie

4、nt computing is proposed that comprises a set of policies for controlling CPU cycles for the mode of local computing, time division between MPT and offloading for the other mode of offloading, and mode selection. Given the CPU-cycle statistics information and channel state information (CSI), the pol

5、icies aim at maximizing the probability of successfully computing given data, called computing probability, under the energy harvesting and deadline constraints. Furthermore, this study reveals that the two simple solutions to achieve the object to support computation load allocation over multiple c

6、hannel realizations, which further increases the computing probability. Last, the two kinds of modes suggest that the feasibility of wirelessly powered mobile cloud computing and the gain of its optimal control. And the future aspect to study is simply to be answer. Key words: wireless power transfe

7、r; energy harvesting communications; mobile cloud computing; energy efficient computingIntroduction Mobile cloud computing (MCC) as an emerging computing paradigm integrates cloud computing and mobile computing to enhance the computation performance of mobile devices. The objective of MCC is to exte

8、nd powerful computing capability of the resource-rich clouds to the resource-constrained mobile devices (e.g., laptop, tablet and smartphone) so as to reduce computation time, conserve local resources, especially battery, and extend storage capacity. To achieve this objective, MCC needs to transfer

9、resource-intensive computations from mobile devices to clouds, referred to as computation offloading. The core of computation offloading is to decide on which computation tasks should be executed on the mobile device or on the cloud, and how to schedule local and cloud resource to implement task off

10、loading. The explosive growth of Internet of Things (IOT) and mobile communication is leading to the deployment of tens of billions of cloud-based mobile sensors and wearable computing devices in near future (Huang & Chae, 2010). Prolonging their battery lives and enhancing their computing capabilit

11、ies are two key design challenges. They can be tackled by two promising technologies: microwave power transfer (MPT) for powering the mobiles computation-intensive tasks from the mobiles to the cloud and mobile computation offloading (MCO). Two technologies are seamlessly integrated in the current w

12、ork to develop a novel design framework for realizing wirelessly powered mobile cloud computing under the criterion of maximizing the probability of successfully computing given data, called computing probability. The framework is feasible since MPT has been proven in various experiments for powerin

13、g small devices such as sensors or even small-scale airplanes and helicopters. Furthermore, sensors and wearable computing devices targeted in the framework are expected to be connected by the cloud- based IOT in the future, providing a suitable platform for realizing MCO.Materials MCO has been an a

14、ctive research area in computer science where research has focused on designing mobile-cloud systems and software architectures, virtual machine migration design in the cloud and code partitioning techniques in the mobiles for reducing the energy consumption and improving the computing performance o

15、f mobiles. Nevertheless, implementation of MCO requires data transmission and message passing over wireless channels, incurring transmission power consumption. The existence of such a tradeoff has motivated cross-disciplinary research on jointly designing MCO and adaptive transmission algorithms to

16、maximize the mobile energy savings. A stochastic control algorithm was proposed for adapting the offloaded components of an application to a time-varying wireless channel. Furthermore, multiuser computation offloading in a multi-cell system was explored by Shinohara (2014), where the radio and compu

17、tational resources were jointly allocated for maximizing the energy savings under the latency constraints.According to Swan (2012), the threshold-based offloading policy was derived for the system with intermittent connectivity between the mobile and cloud. Lastly, the CPU-cycle frequencies are join

18、tly controlled with MCO given a more skilled and increasingly appropriatewireless channel. The framework is further developed in the current work to include the new feature of MPT (Kosta et al., 2012). This introduces several new design challenges. Among others, the algorithmic design of local compu

19、ting and offloading becomes more complex under the energy harvesting constraint due to MPT, which prevents energy consumption from exceeding the amount of harvested energy at every time instant. Another challenge is that MPT and offloading time share the mobile antenna and the time division has to b

20、e optimized.Now the technology is being further developed to power wireless communications. This has resulted in the emergence of an active field called simultaneous wireless information and power transfer (SWIPT). The MPT technology has been developed for point-to-point high power transmission in t

21、he past decades (Brown, 1984). Furthermore, existing wireless networks such as cognitive radio and cellular networks have been redesigned to feature MPT. Most prior work on SWIPT aims at optimizing communication techniques to maximize the MPT efficiency and system throughput. In contrast, the curren

22、t work focuses on optimizing the local computing and offloading under a different design criterion of maximum computing probability (Huang & Lau, 2014).Methods and ResultsConsider a single-user system comprising one multi-antenna base station (BS) using transmit/receive beamforming for transferring

23、power to a single-antenna mobile or relaying offloaded data from the mobile to the cloud. To compute a fixed amount of data, the mobile operates in one of the two available modes: Local computing and offloading: in the mode of local computing, MPT occurs simultaneously as computing based on the cont

24、rollable CPU-cycle frequencies. Nevertheless, in the mode of offloading, the given computation duration is adaptively partitioned for separate MPT and offloading since they share the mobile antenna (Shinohara, 2014). Assume that the mobile has the knowledge of statistics information of CPU cycles an

25、d channel state information (CSI). The individual modes as well as mode selection are optimized for maximizing the computing probability under the energy harvesting and deadline constraints. For tractability, the metric is transformed into equivalent ones, namely average mobile energy consumption an

26、d mobile energy savings, for the modes of local computing and offloading, respectively. Compared with the prior work, the current work integrates MPT with the mobile cloud computing, which introduces new theoretical challenges. In particular, the energy harvesting constraint arising from MPT makes t

27、he optimization problem for local computing non-convex. To tackle the challenge, the convex relaxation technique is applied without compromising the optimality of the solution. It is shown in the sequel that the local computing policy is a special case of the current work where the transferred power

28、 is sufficiently high by Swan (2012). Furthermore, the case of dynamic channel for mobile cloud computing is explored. Approximation methods are used for deriving the simple and close-to-optimal policies.Mobile mode selection: The above results are combined to select the mobile mode for maximizing t

29、he computing probability. Given feasible computing in both modes, the only oneyielding the larger energy savings is preferred and the selection criterion is derived in terms of thresholds on the BS transmission power as well as the deadline for computing (Huang et al., 2012).Optimal data allocation

30、for a dynamic channel: Last, the above results are extended to the case of a dynamic channel, modeled as independent and identically distributed. block fading, and non-causal CSI at the mobile (acquired from e.g., channel prediction). The problem of optimizing an individual mobile mode (local comput

31、ing or offloading) is formulated based on the master-and-slave model using the same metric as the fixed-channel counterpart (Kumar & Liu, 2013). ConclusionWireless and mobile computing technologies provide more possibilities for accessing services conveniently. Mobile devices will be improved in ter

32、ms of power, CPU, and storage. Mobile cloud computing has emerged as a new paradigm and extension of cloud computing.By two kinds of available modes, we can purely know of the Energy Efcient Mobile Cloud Computing. Through my study for the Mobile Cloud Computing, we are here exposing two simple solu

33、tions to solve this problem. Although my research is pretty basic, it still benefit the process of the development for mobile cloud computing and how to make it energy efficient. We believe that exploring other alternatives, such as introducing a middleware based architecture using an optimizing offloading algorithm