Active and Dynamic Beam Tracking UnderStochastic Mobility

TL;DR

This paper introduces an active beam tracking algorithm for UAV-to-UAV mmWave communications that adaptively learns the angle of arrival using Bayesian methods, outperforming existing strategies especially under unpredictable movements.

Contribution

The paper presents a novel Bayesian-based active learning algorithm for beam tracking that adapts to unpredictable UAV movements, improving accuracy and efficiency.

Findings

Significant performance improvements over existing strategies.

Effective adaptive pilot allocation balancing mutual information and spectral efficiency.

Robustness to unpredictable UAV movements.

Abstract

We consider the problem of active and sequential beam tracking at mmWave frequencies and above. We focus on the dynamic scenario of a UAV to UAV communications where we formulate the problem to be equivalent to tracking an optimal beamforming vector along the line-of-sight path. In this setting, the resulting beam ideally points in the direction of the angle of arrival with sufficiently high resolution. Existing solutions account for predictable movements or small random movements using filtering strategies or by accounting for predictable mobility but must resort to re-estimation protocols when tracking fails due to unpredictable movements. We propose an algorithm for active learning of the AoA through evolving a Bayesian posterior probability belief which is utilized for a sequential selection of beamforming vectors. We propose an adaptive pilot allocation strategy based on a trade-off of mutual information versus spectral efficiency. Numerically, we analyze the performance of our proposed algorithm and demonstrate significant improvements over existing strategies.