Abstract
Multi-access edge computing (MEC) is a key technology to support computing-intensive applications in 5G networks. By deploying powerful servers at the edge of wireless networks, MEC can expand the computing power of mobile devices by offloading computationally intensive tasks to MEC servers. This paper considers a multi-user MEC wireless network, where multiple mobile devices can associate and perform computation offload to a MEC server connected to a base station (BS) through a wireless channel. Whether the computing task is executed locally on the user device or offloaded to the MEC server, this decision should adapt to the time-varying network dynamics. Considering the dynamic nature of the environment, this paper proposes a deep reinforcement learning-based approach to the formulation of a non-convex problem that minimizes computational cost in terms of total latency.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call