Measurement-Driven Learning-Based Beam Selection for Hybrid Beamforming at 26.5 GHz

2026-06-29

AI Summary

This work addresses the excessive overhead of exhaustive beam search in indoor millimeter-wave hybrid beamforming by formulating beam selection as a supervised learning task based on real-world channel measurements at 26.5 GHz. Two efficient approaches are proposed: one leveraging geometric features through a deep neural network, and another that operates without location information using only a small number of pilot signals. By integrating synchronized software-defined radio (SDR) platforms with wideband channel sounding, the proposed methods significantly reduce training overhead in an office corridor environment while achieving beam prediction accuracy close to the SNR-optimal benchmark. This study represents the first successful integration of real-environment channel measurements with learning-driven beam selection.

Abstract

This paper investigates learning-assisted transmit beam selection for indoor millimeter-wave (mmWave) systems operating with hybrid beamforming and joint transmission. A synchronized SDR-based testbed at 26.5 GHz band is deployed to collect wideband channel measurements in a realistic office corridor environment. Using the measurement dataset, beam selection is formulated as a supervised learning problem aiming to approximate the SNR-optimal beam obtained through exhaustive sweeping. Two complementary approaches are examined: a geometry-driven Deep Neural Network (DNN) that predicts the optimal beam from spatial features, and a pilots-only method that infers suitable beams using a limited number of sounded pilot beams without positional information. Experimental results demonstrate high prediction accuracy and significant reduction in beam search overhead compared to exhaustive sweeping, highlighting the effectiveness of measurement-driven learning for practical indoor mmWave beam management.

Problem

• beam selection
• hybrid beamforming
• millimeter-wave
• learning-based
• SNR-optimal

Innovation

• measurement-driven learning
• hybrid beamforming
• mmWave beam selection
• deep neural network
• pilot-based inference