Cache‐enabled reinforcement learning scheme for power allocation and user selection in opportunistic downlink NOMA transmissions

Abstract

Non-orthogonal multiple access (NOMA) allows multiple user equipment (UE) to simultaneously share the same resource blocks using varying levels of transmit power at the base station (BS) side. Proper allocation of transmission power and selection of candidate users for pairing over the same resource block are critical for an efficient utilization of the available resources. Optimal UE selection and power splitting among paired UEs can be made through an exhaustive search over the space of all possible solutions. However, the cost incurred by such approach can render it practically infeasible. Reinforcement learning (RL) deploying double deep-Q networks (DDQN) is a promising framework that can be adopted for tackling the problem. In this article, an RL-based DDQN scheme is proposed for user pairing in opportunistic access to downlink NOMA systems with capacity-limited backhaul link connections. The proposed algorithm relies on proactive data caching to alleviate the throttling caused by backhaul bottlenecks, and optimized UE selection and power allocation are accomplished through the continuous interaction between an RL agent and the NOMA environment to increase the overall system throughput. Simulation results are presented to showcase the near-optimal strategy achieved by the proposed scheme.

Energy-Efficient Trajectory Planning With Joint Device Selection and Power Splitting for mmWaves-Enabled UAV-NOMA Networks

“…Pairing NOMA receiving devices optimally for downlink transmissions generally requires a complete scan exhausting all possible groupings. However the computational burden of such a brute force approach is huge which deems the application of the optimal solver unrealistic from a practical standpoint [32]. Numerous strategies were devised for the appropriate grouping of receiving devices over unified NOMA RBs.…”

“…In addition, it is of critical importance to optimize the continuous adaptation of various allocated power portions within a maximum allowable budget of available transmission power as well as the dynamic activation of the receiving devices to reap as much of the promised performance of NOMA-based operation as possible [30], [31]. Moreover, making proper choices regarding the selection of appropriate receiving devices to add to a certain NOMA message stack is important to attain boosted sum-rate levels [32]. In addition, each time the UAV moves position, power allocation and device pairing need to be re-optimized, resulting in a surge in complexity and energy consumption.…”