Delay Minimization for NOMA-MEC Offloading in ABS-Aided Maritime Communication Networks

As maritime activities increase, the demand for high-speed, low-latency communications and efficient task processing in maritime communication networks (MCNets) intensifies. Currently, sea-surface communications and tasks mainly depend on on-shore base stations, which have limited coverage, and marine satellites, which offer broader reach but cannot provide high-speed connections and are costly. Motivated by the above, we investigate the deployment of air base stations (ABSs) in MCNets to offer flexible coverage and handle maritime-specific tasks. ABS-aided MCNets benefit from the integration of mobile edge computing (MEC) and non-orthogonal multiple access (NOMA) techniques. Specifically, by optimizing the task offloading and resource allocation, we formulate a total task processing delay minimization problem under the constraints of task time-to-live (TTL) and successive interference cancellation (SIC) decoding threshold. Since the formulated problem is non-convex, we decouple it and solve it iteratively. The channel allocation, power control, offloading ratio, and computation resource allocation are tackled in turn. Simulation results demonstrate that the proposed scheme significantly outperforms existing schemes in reducing the total task processing delay. Notably, the incorporation of NOMA markedly enhances the task offloading rate, further decreasing the total task processing delay. The proposed scheme cuts the total task processing delay by at least 7.7% compared to the state-of-the-art schemes and by 40.9% against the local computing benchmark. Additionally, the performance of the proposed scheme, including its complexity, convergence, and the performance loss due to iterative optimization, is analyzed.