Mobility-Aware Offloading and Resource Allocation in NOMA-MEC Systems via DC

Multi-access edge computing (MEC) is a promising computing paradigm that incorporates computing capability into a radio access network (RAN) nearby user equipment (UE) to improve the quality of experience. However, MEC also encounters many challenges inherent in RANs, such as mobility management. In this letter, we develop a novel computation offloading scheme utilizing non-orthogonal multiple access (NOMA) and dual connectivity (DC) and focus on jointly optimizing task segmentation and power allocation to minimize the total energy consumption. Specifically, the joint optimization problem is formulated as a mixed-integer nonlinear programming (MINLP) problem that is non-deterministic polynomial hard (NP-hard). To solve this problem, we propose a new heuristic algorithm to obtain a sub-optimal solution. We then propose an online learning algorithm based on a twin delayed deep deterministic policy gradient (TD3) to meet user mobility requirements. The numerical results show that the proposed scheme outperforms other schemes, and the TD3-based algorithm has comparable energy consumption performance and dramatically reduces the execution time compared with the heuristic algorithm.

Automated summary

Multi-access edge computing (MEC) is a computing paradigm that integrates computing capability into a wireless access network to enhance user experience. This letter proposes a novel computation offloading scheme using non-orthogonal multiple access (NOMA) and dual connectivity (DC) to minimize energy consumption. The scheme is optimized by jointly optimizing task segmentation and power allocation, and the optimization problem is solved using a heuristic algorithm and an online learning algorithm based on twin delayed deep deterministic policy gradient (TD3).