Dynamic Priority-Based Computation Offloading for Integrated Maritime-Satellite Mobile Networks

In order to meet the increasing demand for delay-sensitive and computing-intensive applications of maritime users, we first propose an integrated satellite-maritime mobile edge computing framework. The framework considers a maritime mobile communication network consisting of maritime satellites and shipborne base stations, and maritime computation offloading coordinated by the terrestrial cloud and the shipborne edge servers. Secondly, we dynamically calculate the priority that characterizes the urgency of offloading tasks with reinforcement learning. Based on the dynamic priority, a maritime computation offloading method is proposed to optimize the system cost. Finally, simulation results verify the effectiveness and convergence of the proposed method in terms of offloading delay, user energy consumption, offloading response rate and average offloading cost.