Energy Efficiency Maximization in Cooperative Hybrid VLC/RF Networks with NOMA

In this paper, a cooperative hybrid visible light communications (VLC)/radio frequency (RF) network that employs non-orthogonal multiple access (NOMA) is investigated. More specifically, the resource allocation and the operating mode of the system is optimized to maximize a weighted energy efficiency metric, taking into accounts the particularities of the hybrid network. The resulting optimization problem is efficiently tackled with the use of Dinkelbach's algorithm and difference of convex programming. Using this solution, a deep neural network (DNN) is trained to find the operation mode and alleviate the computational cost of the overall problem leading to a close solution to the optimal. Finally, the effectiveness of the proposed setup is presented via Monte Carlo simulation results.