Two-Sided Learning for NOMA-Based Random Access in IoT Networks

In the Internet-of-Things (IoT), different types of devices can co-exist within a network. For example, there be cheap but inflexible and flexible in terms radio frequency (RF) capabilities. Thus, order to support ways improve throughput, we propose multichannel random access scheme based on power-domain non-orthogonal multiple (NOMA), where each or dynamic device (DD) dynamically choose one channels when it has packet send. addition, since DDs need learn channel selection probabilities maximize throughput DDs, consider two-sided learning multi-armed bandit (MAB) formulation rewards are decided by outcomes at base station (BS) speed DDs. Simulation results confirm that help allows proposed NOMA-based approach achieve near maximum throughput.