Energy Efficient Resource Allocation in Downlink MC-NOMA Networks With Finite Alphabet Inputs

Resource allocation in multi-carrier non-orthogonal multiple access (MC-NOMA) systems typically assumes Gaussian signaling inputs. However, practical communication systems employ finite alphabet inputs, rendering Gaussian-based resource allocation suboptimal. Motivated by this practical limitation, we propose an energy-efficient power and subchannel allocation strategy specifically tailored for MC-NOMA networks with finite alphabet inputs. The resulting optimization problem is inherently non-convex; hence, we approach it by decomposing it into two sequential subproblems, enabling us to propose an efficient algorithm that first allocates subchannels and subsequently optimizes the power allocation for each user. The power allocation subproblem is transformed into convex form through the application of Jensen’s inequality, successive pseudo-concave approximation, and Dinkelbach’s algorithm. Simulation results demonstrate that the proposed approach significantly improves energy efficiency, providing a more accurate representation of practical system performance compared to traditional Gaussian input-based resource allocation schemes.