This paper introduces the Space Splitting-based User Selection (SS-US) algorithm, a novel low-complexity and massively parallelizable method for MU-MIMO user selection. It addresses the scalability challenges posed by the combinatorial nature of existing approaches in dense, uplink-oriented, and latency-critical AI-native traffic scenarios, while achieving comparable spectral efficiency to state-of-the-art baselines.

This paper explores the integration of large Artificial Intelligence (AI) models into future wireless communication systems, addressing the increasing complexity and demands of next-generation networks. It proposes an architecture for these models, highlights their benefits in data analysis, resource allocation, and real-time adaptation, and discusses significant challenges alongside potential solutions.