Task offloading and resource allocation for edge-enabled mobile learning

Abstract

Mobile learning has evolved into a new format of education based on communication and computer technology that is favored by an increasing number of learning users thanks to the development of wireless communication networks, mobile edge computing, artificial intelligence, and mobile devices. However, due to the constrained data processing capacity of mobile devices, efficient and effective interactive mobile learning is a challenge. Therefore, for mobile learning, we propose a "Cloud, Edge and End" fusion system architecture. Through task offloading and resource allocation for edge-enabled mobile learning to reduce the time and energy consumption of user equipment. Then, we present the proposed solutions that uses the minimum cost maximum flow (MCMF) algorithm to deal with the offloading problem and the deep Q network (DQN) algorithm to deal with the resource allocation problem respectively. Finally, the performance evaluation shows that the proposed offloading and resource allocation scheme can improve system performance, save energy, and satisfy the needs of learning users.