Device Scheduling and Computation Offloading in Mobile Edge Computing Networks: A Novel NOMA Scheme

Massive devices in Internet of Things (IoT) typically have limited resources, and thus it is hard to satisfy the computation demands of various IoT applications by local computation of the collected data at the devices. Mobile edge computing (MEC) tackles this challenge by acquiring and computing the collected data at the network edge. In this paper, a novel non-orthogonal multiple access (NOMA) scheme that combines the merits of the time-division multiple access (TDMA) and the conventional NOMA is proposed for data collection and offloading in MEC networks. Specifically, the proposed novel NOMA scheme takes advantage of the strengths of both TDMA and conventional NOMA by allowing only a subset of devices to offload data and the remaining devices to collect data simultaneously. Based on the proposed novel NOMA scheme, optimal and low-complexity suboptimal solutions to the total offloading throughput maximization problem are proposed. Simulation results demonstrate the superiority of the proposed novel NOMA scheme compared to the TDMA, the frequency-division multiple access (FDMA) and the conventional NOMA schemes.