ML Based User Pairing Scheme for Integrated NOMA-RAW Channel Access Mechanism for IEEE 802.11ah Networks

Non-Orthogonal Multiple Access (NOMA) is a key technology for IEEE 802.11ah-based IoT networks, enabling efficient use of spectral resources, low power consumption, scalability, and fairness in the network. User pairing is a crucial technique in NOMA because it helps to achieve a better balance of resources and improve overall network performance. In this paper, we present an ML-based user pairing (MLUP) approach for dense Internet of Things networks that comply with IEEE 802.11ah. The approach utilizes NOMA at the physical layer and the Restricted Access Window (RAW) mechanism at the MAC layer. The throughput of the NOMA-RAW approach under the proposed MLUP mechanism is evaluated using an analytical model. The findings demonstrate that the proposed pairing scheme considerably improves the performance of the IEEE 802.11ah network. Finally, we validate the analytical results by using ns-3 simulations.