Effective Computational Efficiency Maximization in Cooperative NOMA Based MEC System.

The paper mainly studies how to better apply non-orthogonal multiple access (NOMA) to mobile edge computing (MEC) system under the condition of considering offload interruption, and adopts user collaboration(UC) and wireless power transfer (WPT) technology. First, two users use WPT to collect energy from the base station equipped with the MEC server. Then, to better increase the system capacity and performance, the near user acts as a relay to help the far user offload some of the tasks to the MEC server, using the NOMA protocol for both users. To obtain an energy-efficient MEC design, we first derive the outage probability for both users, then the offloading time allocation and power allocation are jointly designed to maximize the effective computational efficiency, which is a non-convex problem. To deal with the problem, we first transform it into a relaxation problem and then solve it using the sequential convex approximation (SCA) method. Finally, numerical results demonstrate that the proposed scheme introduces NOMA and UC into MEC simultaneously, which has better advantages compared with other schemes.