Energy-efficient and delay-aware mobile cloud offloading over cellular networks

摘要

The latest mobile applications, such as GPS, games, virus scanning, and face detection and recognition, are compute-intensive applications consuming a lot of energy when executed, causing the limited-capacity batteries of the mobile devices to drain rapidly. Thus, the fifth generation standard has proposed the deployment of cloud radio access networks (C-RANs). By deploying C-RANs, the users can conserve their mobiles' limited battery energy by running the compute-intensive and energy-hungry tasks or apps on the cloud. However, offloading the tasks to the cloud might not be always better, in terms of energy conservation, than running the task on the device. Many factors should be considered, such as the offloading delay, wireless channel conditions, and the energy consumption at the cloud. Therefore, in this paper, we propose an algorithm to offload the tasks to the cloud over a cellular network, with an objective of achieving an energy-efficient offloading while considering the execution time of the tasks. The algorithm takes into account the wireless channel conditions, the load of the servers in the cloud, and the mobility of the users. The results obtained show that the proposed algorithm saves the transmission and execution energies while satisfying the delay constraints of various tasks.