Near-Field Multiuser Beam Training for XL-MIMO: An End-to-End Interference-Aware Approach with Pilot Limitations

TL;DR

This paper introduces a deep learning framework for near-field multiuser beam training in XL-MIMO systems, efficiently predicting beam indices with limited pilots while accounting for interference.

Contribution

It proposes an end-to-end interference-aware deep learning method that directly predicts beam indices from sensing measurements, reducing pilot overhead in XL-MIMO near-field scenarios.

Findings

Achieves near-optimal sum-rate performance.

Provides higher throughput under pilot constraints.

Effectively accounts for multiuser interference.

Abstract

Near-field propagation in extremely large-scale MIMO (XL-MIMO) enlarges the beam training (BT) search space by introducing an additional range dimension, which makes conventional codebook-based beam sweeping prohibitively expensive under limited pilot resources, especially for multiuser sub-connected hybrid architectures. This letter proposes a deep-learning-based interference-aware multiuser BT framework (DL-IABT) that directly predicts analog beam indices from a small number of uplink sensing measurements. By exploiting a subarray-level approximation, a far-field codebook is adopted to represent each subarray response with negligible mismatch. To enable end-to-end (E2E) learning, we derive a variant-MSE surrogate loss by eliminating the digital precoder through a closed-form MMSE solution from KKT conditions, which implicitly accounts for multiuser interference (MUI). The proposed network integrates a complex-valued sensing front-end, a shared complex-valued encoder, a Transformer-based multiuser predictor, and a scalable Gumbel--Softmax beam selection head. Simulation results show that DL-IABT achieves near-optimal sum-rate performance while providing markedly higher effective throughput under pilot overhead constraints.