The deployment of 5G networks has incorporated advanced multiple access technologies like sparse code multiple access (SCMA) to address growing demands for high-speed connectivity and massive device access. As a Non-Orthogonal Multiple Access technique, SCMA enables multiple users to share identical time-frequency resources through sparse codebook-based multiplexing. Nevertheless, achieving efficient scheduling in SCMA networks remains challenging due to the inherent complexities in dynamic resource allocation. This paper proposed two artificial intelligence-based approaches for resource scheduling in 5G SCMA networks: a multi-agent deep reinforcement learning (MARL)-based approach and a large language model (LLM)-empowered methodology. We systematically investigate these AI techniques to develop adaptive resource scheduling policies capable of responding to diverse network conditions. Simulation results validate that the proposed MARL-based and LLM-based schedulers not only effectively learn optimal scheduling strategies but also outperform conventional algorithms, particularly in terms of system throughput and user fairness metrics.
Zhao et al. (Tue,) studied this question.