Digital Twin-Enabled Computation Offloading in UAV-Assisted MEC Emergency Networks

With the advantage of flexible deployment in all terrains, unmanned aerial vehicles (UAV) equipped with lightweight computing module can bring mobile edge computing (MEC) service closer to both survivors and rescuers in disaster scenarios. However, the mismatch between the sudden computing demands of mobile users (MUs) and limited computation resources of UAVs cannot meet the high delay sensitivity requirements for computation offloading. In this letter, we aim to exploit the real-time prediction capability of digital twin (DT) to optimize the offloading decision under uncertainty. Considering the mapping deviations between the DT and the real network status, we formulate the computation offloading problem as an online matching problem in uncertain bipartite graph. Furthermore, we transform this matching problem into a multi-armed bandits (MAB) problem and finally solve it through an upper confidence bound (UCB)- based stable matching algorithm. The simulation results prove the superiority of our proposed method.

Automated summary

In this paper, a method is proposed to optimize mobile edge computing using digital twin, and to make offloading decisions based on real-time predictions under uncertainty.