期刊名称：International Journal of Advances in Engineering and Management
电子版ISSN：2395-5252
出版年度：2021
卷号：3
期号：5
页码：919-923
DOI：10.35629/5252-0305792794
语种：English
出版社：IJAEM JOURNAL
摘要：Nowadays, mobile devices are responsible for processing more and more Computational intensive tasks, such as data processing, artificial intelligence, and virtual reality. Despite the development of mobile devices, these devices may not be able to process all their tasks locally with a low latency due to their limited computational resources. To facilitate efficient task processing, mobile edge computing (MEC), also known as fog computing and multi-accessed gecomputing, is introduced. MEC facilitates mobile devices to offload their computational intensive tasks to near by edge nodes for processing in order to reduce the task processing delay. It can also reduce the ratio of dropped tasks for those delay - sensitive tasks. In MEC, there are two main questions related to task offloading. The first question is whether a mobile device should offload its task to an edge node or not. These condquestion is that if a mobile device decides toper form offloading, then which edge node should the device offload its task to. In this work, we consider non - divisible and delay - sensitive tasks as well as edge load dynamics, and formulate a task offloading problem to minimize the expected long-term cost. We propose a model - free deep reinforcement learning - based distributed algorithm, where each device an determine its offloading decision without knowing the task models and offloading decision of other devices. To improve the estimation of the long - term cost in the algorithm, we incorporate the long short - term memory (LSTM), dueling deep Q - network (DQN), and double DQN techniques. Simulation results with 50 mobile devices and five edge nodes show that the proposed algorithm can reduce the ratio of dropped tasks and average task delay by 86.4% − 95.4% and 18.0% −30.1%, respectively, when compared with several existing algorithms.
