Efficient, Adaptive Near-Field Beam Training based on Linear Bandit

TL;DR

This paper introduces a linear bandit-based adaptive beam training framework for near-field communication that significantly reduces pilot overhead and improves SNR by leveraging Thompson Sampling and spatial correlation modeling.

Contribution

It presents a novel adaptive beam training method using linear bandits with correlated priors, achieving high efficiency and near-optimal performance in multipath near-field scenarios.

Findings

Reduces pilot overhead by up to 90%.

Achieves over 2dB SNR gain compared to baselines.

Approaches full-CSI performance with unconstrained pilot overhead.

Abstract

This letter proposes a linear bandit-based beam training framework for near-field communication under multi-path channels. By leveraging Thompson Sampling (TS), the framework adaptively balances exploration and exploitation to maximize cumulative beamforming gain under limited pilot overhead. To ensure data-efficient learning, we incorporate a correlated Gaussian prior in the DFT domain, using a Gaussian kernel to capture spatial correlations and near-field energy leakage. We develop three TS strategies: codebook-constrained search for rapid convergence via structural regularization, continuous-space search to achieve near-optimal performance, and a two-stage hybrid refinement scheme that balances convergence speed and estimation accuracy. Simulation results show that the proposed framework reduces pilot overhead by up to 90\% while achieving more than a 2dB SNR gain over baselines in multipath environments. Furthermore, the continuous-space search is shown to be asymptotically optimal, approaching the full-CSI bound when the pilot overhead is unconstrained.