Low Overhead Beam Alignment for Mobile Millimeter Channel Based on Continuous-Time Prediction

TL;DR

This paper tackles the high overhead of maintaining directional gain in mobile mmWave links by introducing Adaptive Online Beam Alignment (AOBA), which combines low-overhead beam prediction over a small candidate set with periodic full-beam scanning. AOBA uses an ODE-LSTM-based beam predictor to estimate the optimal beam index $\\hat{q}^*(t)$ during pilot-free intervals, while three candidate-beam selection strategies and two mode-switching rules manage when to switch between prediction and scanning. The approach significantly reduces training overhead—by roughly 48% in simulations—while preserving near-upper-bound beamforming gains, and demonstrates robustness to varying UE speeds and longer prediction horizons. These results suggest AOBA offers practical, scalable improvements for mobile mmWave deployments and can be extended to multi-user scenarios in future work.

Abstract

In millimeter-wave (mmWave) communications, directional transmission based on beamforming is important to compensate for high pathloss. To maintain the desired direction transmission gain, beam scanning that involves the transmitter sending the pilot signal over all available beam directions to find the optimal beam is often considered. Alternatively, beam tracking using partial beams can save the beam training overhead through algorithms such as statistical analysis models and kalman filter (KF). Unfortunately, existing beam tracking solutions are limited to a fixed beam variation pattern. In this work, we propose an adaptive online beam alignment (AOBA) scheme, which aims to reduce training overhead and achieve accurate beam alignment for any movement profile. The proposed AOBA periodically performs beam tracking using a small amount but carefully selected candidate beams and switches to beam scanning using all available beams based on a given switching rule. During the interval without the pilot signal, the optimal beam at an arbitrary time instant is predicted with the aid of the recently proposed ordinary differential equation (ODE)-long short-term memory (LSTM) model. Extensive simulations are conducted to evaluate the performance of the proposed AOBA in comparison with several existing beam alignment schemes.

Figures (6)

• Figure 1: Time-domain operation of the proposed beam alignment scheme.
• Figure 2: Illustration of three beam selection strategies.
• Figure 3: The workflow of the proposed AOBA, where $t_{n,i}$ represent the $i$-th prediction instant between $t_n$ and $t_{n+1}$.
• Figure 4: Comparison of prediction accuracy of three candidate beam selection strategies versus UE velocity.
• Figure 5: Comparison of different beam prediction methods for 1-second trajectories.