UAV-Assisted Multi-Access Computation Offloading via Hybrid NOMA and FDMA in Marine Networks

With the rapid development of marine networks, there have been growing demands for computation-intensive and delay-sensitive marine applications and services. However, the limited underwater energy supply and the acoustic channels result in the low efficiency for computing tasks and high transmission delay. In this paper, we investigate the unmanned aerial vehicles (UAVs)-assisted multi-access computation offloading in marine networks, with the objective of minimizing the energy consumption of ocean devices. Specifically, for the underwater segment, we consider the scenario that multiple underwater sensor nodes (USNs) covered by the unmanned surface vehicle (USV) upload their sensing data via non-orthogonal multiple access (NOMA) for improving the channel utilization. For the radio frequency segment, we consider the scenario that multiple UAVs hovering in the air act as the aerial edge servers for providing computing services, in which the USV offloads the workloads to UAVs via frequency division multiple access (FDMA) for avoiding their co-channel interference, while taking into account that a malicious node overhears the USV's offloading transmission. To improve the computation offloading efficiency, we formulate an optimization problem for USNs and USV to minimize the total energy consumption by jointly optimizing the USN's uploading time, USV's computation offloading, USV's offloading time, and the secrecy provisioning. Despite the non-convexity of the formulated joint optimization problem, we exploit a layered structure to decompose the problem, and propose efficient algorithms to obtain the optimal solutions. Finally, we conduct simulations to validate the effectiveness and efficiency of the proposed algorithms. Numerical results demonstrate that our algorithms can significantly reduce the energy consumption in comparison with the benchmark schemes.