Novel Channel Aware Strategies for Power Control in a Downlink NOMA Network With Mode Switching and User Selection

This paper considers a multi-user downlink non-orthogonal multiple access (NOMA) framework comprising of an access point (AP), a cluster of N near-users (NUs), and a cluster of F far-users (FUs). The AP shares direct links to both NUs and FUs. Using only a one-bit feedback from the selected users, the network switches between NOMA, cooperative NOMA, and orthogonal multiple access (OMA) modes. Considering channel-aware dynamic power allocation (DPA), closed-form expressions are derived for the outage probability, throughput, and EE for the first time in literature with novel power control, mode selection, and user selection. Further, it is demonstrated that the power allocation with very limited channel state information (CSI) results in a significant gain in throughput and also saves power at AP, thereby increasing EE. We show in that intelligent user selection and use of channel-aware techniques is the key to high spectral efficiency (SE) and energy efficiency (EE). It is shown how the minimum transmit power at the AP and the NU, as well as the best NOMA power-allocation can be determined based on limited channel knowledge to maximize SE and EE. The use of minimum power needed for all three modes further improve the EE of DPA by 16.43%. We also demonstrate that in contrast to the fixed power allocation (FPA) scheme, the DPA scheme increases the mode probability of NM, thereby increasing SE. Closed-form expressions are derived for the minimum average power consumption and power increments to switch to modes yielding larger throughput. We demonstrate that choosing the optimal target rate is essential to maximize the EE. Monte Carlo simulations validate the derived expressions.