Deep Learning based Efficient User Association and Subchannel Assignment in Multi-Cell Multi-Carrier NOMA Networks

In this paper, we investigate the problem of efficient user association and subchannel assignment in multi-cell multi-carrier nonorthogonal multiple access networks. An optimization problem is formulated to maximize the energy efficiency. The proposed solution includes two stages. First, we propose an iterative low complexity algorithm to handle the optimization problem. Second, we use the output from the iterative solution to train two deep neural network (DNN) models. To obtain the final output, the output of the DNNs are mapped in matrix which represents the final user association and subchannel assignment. The simulation results have shown the efficiency of the proposed framework compared to two step 2D matching algorithm and fractional programming (FP) method.