Beam alignment is a critical bottleneck in millimeter wave communication. An ideal beam alignment technique should achieve high beamforming gain with low latency, scale well to systems with higher carrier frequencies, larger antenna arrays and multiple user equipment, and not require hard-to-obtain context information. These qualities are collectively lacking in existing methods. We depart from the conventional codebook-based (CB) approach where the optimal beam is chosen from quantized codebooks and instead propose a grid-free beam alignment method that directly synthesizes the transmit and receive beams from the continuous search space using measurements from a few site-specific probing beams found via a deep learning pipeline. In realistic settings, the proposed method achieves a far superior signal-to-noise ratio (SNR)-latency trade-off compared to the CB baselines: it aligns near-optimal beams 100x faster or equivalently finds beams with 10–15 dB higher average SNR in the same number of searches, relative to an exhaustive search over a conventional codebook.

Beam alignment - finding optimal analog beam-forming (BF) weights - is a critical bottleneck for millimeter wave (mmWave) systems. Existing beam alignment approaches typically assume that devices adopt codebooks of analog beams with uniform coverage, from which a good beam pair is selected after an exhaustive search or sweeping a few candidate beams. In this work, we propose a beam alignment method that is grid-less - the analog beam is synthesized from the continuous set instead of being chosen from a quantized codebook, and one-shot - near-optimal BF weights are directly predicted without searching even a small number of candidates. With unsupervised training, the proposed method uses a few learned probing beams to sense the channel and predict the BF weights. Our experiments show that it can get within 0.32 dB of the hard theoretical upper bound, outperforms the exhaustive search in terms of the signal-to-noise ratio (SNR), reduces the beam sweeping latency by over 20×, while scaling optimally to multiple UEs and fitting within the 5G NR beam alignment framework.

Beam alignment – the process of finding an optimal directional beam pair – is a challenging procedure crucial to millimeter wave (mmWave) communication systems. We propose a novel beam alignment method that learns a site-specific probing codebook and uses the probing codebook measurements to predict the optimal narrow beam. An end-to-end neural network (NN) architecture is designed to jointly learn the probing codebook and the beam predictor. The learned codebook consists of site-specific probing beams that can capture particular characteristics of the propagation environment. The proposed method relies on beam sweeping of the learned probing codebook, does not require additional context information, and is compatible with the beam sweeping-based beam alignment framework in 5G. Using realistic ray-tracing datasets, we demonstrate that the proposed method can achieve high beam alignment accuracy and signal-to-noise ratio (SNR) while significantly – by roughly a factor of 3 in our setting – reducing the beam sweeping complexity and latency.

Beam alignment - the process of finding an optimal directional beam pair - is a challenging procedure crucial to millimeter wave (mmWave) communication systems. In this work, we propose a beam alignment method that learns a site-specific probing codebook and uses the probing codebook measurements to predict the optimal narrow beam. A novel neural network (NN) architecture is designed to jointly learn the probing codebook and the beam predictor in an end-to-end fashion. The learned codebook consists of site-specific probing beams that can capture particular characteristics of the propagation environment. The proposed method relies on beam sweeping of the learned probing codebook, does not require additional context information and is compatible with the beam sweeping-based beam alignment framework in 5G. We demonstrate using realistic ray-tracing data that the proposed method can achieve high beam alignment accuracy and signal-to-noise ratio (SNR) while significantly reducing the beam sweeping complexity and latency.

Beam alignment is a challenging and time-consuming process for millimeter wave (mmWave) initial access (IA). We propose a beam training method that is assisted by machine learning (ML), where we train ML models to predict the optimal Access Point (AP) and optimal beam for a user equipment (UE) given its Global Positioning System (GPS) coordinates. After a (possibly offline) training phase during which exhaustive or hierarchical beam training is performed, our beam training method predicts a few candidate APs and beams knowing only the location of the UE. We train the models and evaluate the performance with realistic mmWave beamforming (BF) data generated from state-of-the- art ray tracing software. We show that even with dynamic scatterers and imperfect knowledge of the UE locations, our beam training method can reliably find the optimal AP and the optimal beam for a UE while reducing the search time by 4x for AP selection and over 10x for beam selection.