Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Schott, Aron, Acikgöz, Berk, Massoud, Omar, Petrova, Marina, Simić, Ljiljana
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2511.17122
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908668911419392
author Schott, Aron
Acikgöz, Berk
Massoud, Omar
Petrova, Marina
Simić, Ljiljana
author_facet Schott, Aron
Acikgöz, Berk
Massoud, Omar
Petrova, Marina
Simić, Ljiljana
contents Large antenna arrays and beamforming techniques are key components for exploiting the spectrum-rich FR2 bands in next-generation mobile communication networks. Given the site-specific spatio-temporal variations of the mm-wave channel, non-RF sensor inputs and environment awareness can be leveraged to greatly enhance beam management decisions, e.g. via machine learning (ML) techniques. However, the current literature lacks open platforms to gather datasets for the training of such ML techniques and to evaluate novel beam management approaches in real-time, real-world scenarios and full-stack endto-end networks. In this work, we present our SDR-based experimental platform based on OpenAirInterface and are the first to integrate popular low-cost antenna array transceivers, beam sweeping capabilities, and a highly-modular sensor framework and associated interfaces into such a full-stack experimental platform. This enables beam management experimentation in real-world, real-time scenarios and facilitates gathering datasets necessary for developing ML-based beam management protocols that incorporate environment awareness via sensor modalities.
format Preprint
id arxiv_https___arxiv_org_abs_2511_17122
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Unleashing Sensor-Aided Environment Awareness for Beam Management in Beyond-5G Networks: An OpenAirInterface Experimental Platform
Schott, Aron
Acikgöz, Berk
Massoud, Omar
Petrova, Marina
Simić, Ljiljana
Signal Processing
Large antenna arrays and beamforming techniques are key components for exploiting the spectrum-rich FR2 bands in next-generation mobile communication networks. Given the site-specific spatio-temporal variations of the mm-wave channel, non-RF sensor inputs and environment awareness can be leveraged to greatly enhance beam management decisions, e.g. via machine learning (ML) techniques. However, the current literature lacks open platforms to gather datasets for the training of such ML techniques and to evaluate novel beam management approaches in real-time, real-world scenarios and full-stack endto-end networks. In this work, we present our SDR-based experimental platform based on OpenAirInterface and are the first to integrate popular low-cost antenna array transceivers, beam sweeping capabilities, and a highly-modular sensor framework and associated interfaces into such a full-stack experimental platform. This enables beam management experimentation in real-world, real-time scenarios and facilitates gathering datasets necessary for developing ML-based beam management protocols that incorporate environment awareness via sensor modalities.
title Unleashing Sensor-Aided Environment Awareness for Beam Management in Beyond-5G Networks: An OpenAirInterface Experimental Platform
topic Signal Processing
url https://arxiv.org/abs/2511.17122