Two-Timescale Mobility Management for Multi-Cell Mobile Edge Computing

Mobile edge computing (MEC) is a promising technology to support the latency-critical applications of mobile devices by offloading complex computation tasks to edge servers. However, the mobility of devices yield a great challenge on delivering reliable continuous services, especially for those latency-critical applications. Motivated by the fact that the user's location changes slower than the task arrivals, we propose a two-time-scale mobility management framework by joint service migration and power control. The management design is formulated as a long-term energy minimization problem, subject to the reliability requirement of the latency-critical application. Leveraging the Lyapunov optimization technique, we develop an online two-timescale control algorithm to solve the problem. The simulation results demonstrate that our proposed online algorithm can significantly improve the energy and reliability performance compared to the baselines.