A Scheme for Uplink NOMA Communication with Intelligent Resource Allocation for mMTC Traffic over eMBB Traffic

In this work, a study is conducted on the coexistence of two services, enhanced mobile broadband (eMBB) and massive machine-based communication (mMTC), in a fifth generation (5G) wireless communication system. Coexistence of these two services can be achieved by network slicing and non-orthogonal multiple access (NOMA) with successive interference cancellation (SIC) decoding to improve spectral efficiency, but the block error rate (BLER) of mMTC devices needs to be considered for short packet communications (SPC). In this paper, we combine deep reinforcement learning (DRL) proximal policy optimization (PPO) algorithm to efficiently accommodate mMTC devices into available random access (RA) slots while guaranteeing the throughput of eMBB devices to achieve dynamic scheduling of resources to improve system throughput. The performance of the proposed method is compared with recent solutions in the literature, and simulation results show that the proposed method can guarantee a low outage probability for eMBB services and effectively improve the throughput of mMTC services.