Machine Learning-Based mmWave MIMO Beam Tracking in V2I Scenarios: Algorithms and Datasets

TL;DR

This paper introduces a new multimodal dataset and evaluates machine learning algorithms for millimeter wave MIMO beam tracking in V2I scenarios, demonstrating significant performance improvements with a deep neural network model.

Contribution

It provides a novel public dataset combining images, LIDAR, and GNSS data, and evaluates ML-based beam tracking algorithms, highlighting the effectiveness of a ResNet-LSTM model.

Findings

Deep neural network with ResNet and LSTM outperforms other models

Multimodal dataset enables comprehensive evaluation of data fusion methods

Proposed algorithms reduce beam tracking overhead in V2I scenarios

Abstract

This work investigates the use of machine learning applied to the beam tracking problem in 5G networks and beyond. The goal is to decrease the overhead associated to MIMO millimeter wave beamforming. In comparison to beam selection (also called initial beam acquisition), ML-based beam tracking is less investigated in the literature due to factors such as the lack of comprehensive datasets. One of the contributions of this work is a new public multimodal dataset, which includes images, LIDAR information and GNSS positioning, enabling the evaluation of new data fusion algorithms applied to wireless communications. The work also contributes with an evaluation of the performance of beam tracking algorithms, and associated methodology. When considering as inputs the LIDAR data, the coordinates and the information from previously selected beams, the proposed deep neural network based on ResNet and using LSTM layers, significantly outperformed the other beam tracking models.