A Decomposed Learning Framework for Hybrid Precoding
Learning-based hybrid precoding has emerged as a promising solution for low-latency multi-user transmission in large-scale antenna systems. However, its practical deployment is often hindered by high training overhead, and poor generalization across varying system sizes. These limitations fundamentally stem from the fa...