Location-aware Task Offloading in Mobile Edge Computing

As one of the core technologies of 5G, mobile edge computing (MEC) has attracted wide attention from the academic community in recent years. In the mobile edge computing environment, users can choose to offload tasks to edge servers that are closer to the user. But during task processing, the user is likely to move out of the coverage of the edge server currently serving him, which will result in additional transmission delay. Therefore, achieving low task processing delay becomes increasingly challenging due to user mobility in mobile edge computing environment. In this paper, we aim to address this challenge and propose a location-aware task offloading strategy based on deep reinforcement learning algorithm in a mobile edge computing environment. The proposed strategy jointly considers user mobility and allocation of bandwidth and computing resources to minimize the average delay of tasks completion. Experimental results show that our method reduces the average delay of tasks completion by about 10% compared to algorithms that do not predict user motion trajectories.