A Generalized XGBoost-Based Approach for Camera-Driven Beam Selection

Zahra Zarei; Chung G. Kang

doi:10.1109/CCNC54725.2025.10975896

A Generalized XGBoost-Based Approach for Camera-Driven Beam Selection

Zahra Zarei
, Chung G. Kang

School of Electrical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Incorporating non-RF data in beam selection can speed up decision-making in situations that require a thorough search of all candidate options. Therefore, using camera sensory data at base stations to enhance millimeter-wave (mmWave) beam selection is becoming increasingly popular. This paper introduces a machine learning framework using a robust model along with XGBoost in beam prediction scenarios. Our approach enhances beam selection in both single and multi-candidate real-world vehicle-to-infrastructure (V2I) settings.

Original language	English
Title of host publication	2025 IEEE 22nd Consumer Communications and Networking Conference, CCNC 2025
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9798331508050
DOIs	https://doi.org/10.1109/CCNC54725.2025.10975896
Publication status	Published - 2025
Event	22nd IEEE Consumer Communications and Networking Conference, CCNC 2025 - Las Vegas, United States Duration: 2025 Jan 10 → 2025 Jan 13

Publication series

Name	Proceedings - IEEE Consumer Communications and Networking Conference, CCNC
ISSN (Print)	2331-9860

Conference

Conference	22nd IEEE Consumer Communications and Networking Conference, CCNC 2025
Country/Territory	United States
City	Las Vegas
Period	25/1/10 → 25/1/13

Bibliographical note

Publisher Copyright:
© 2025 IEEE.

Keywords

Beam prediction
Machine learning
V2I
XGBoost

ASJC Scopus subject areas

Artificial Intelligence
Computer Networks and Communications
Computer Vision and Pattern Recognition
Electrical and Electronic Engineering

Access to Document

10.1109/CCNC54725.2025.10975896

Cite this

A Generalized XGBoost-Based Approach for Camera-Driven Beam Selection. / Zarei, Zahra; Kang, Chung G.
2025 IEEE 22nd Consumer Communications and Networking Conference, CCNC 2025. Institute of Electrical and Electronics Engineers Inc., 2025. (Proceedings - IEEE Consumer Communications and Networking Conference, CCNC).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Zarei, Z & Kang, CG 2025, A Generalized XGBoost-Based Approach for Camera-Driven Beam Selection. in 2025 IEEE 22nd Consumer Communications and Networking Conference, CCNC 2025. Proceedings - IEEE Consumer Communications and Networking Conference, CCNC, Institute of Electrical and Electronics Engineers Inc., 22nd IEEE Consumer Communications and Networking Conference, CCNC 2025, Las Vegas, United States, 25/1/10. https://doi.org/10.1109/CCNC54725.2025.10975896

@inproceedings{cf0abdd502ce4f4495b132bda1e41bb5,

title = "A Generalized XGBoost-Based Approach for Camera-Driven Beam Selection",

abstract = "Incorporating non-RF data in beam selection can speed up decision-making in situations that require a thorough search of all candidate options. Therefore, using camera sensory data at base stations to enhance millimeter-wave (mmWave) beam selection is becoming increasingly popular. This paper introduces a machine learning framework using a robust model along with XGBoost in beam prediction scenarios. Our approach enhances beam selection in both single and multi-candidate real-world vehicle-to-infrastructure (V2I) settings.",

keywords = "Beam prediction, Machine learning, V2I, XGBoost",

author = "Zahra Zarei and Kang, \{Chung G.\}",

note = "Publisher Copyright: {\textcopyright} 2025 IEEE.; 22nd IEEE Consumer Communications and Networking Conference, CCNC 2025 ; Conference date: 10-01-2025 Through 13-01-2025",

year = "2025",

doi = "10.1109/CCNC54725.2025.10975896",

language = "English",

series = "Proceedings - IEEE Consumer Communications and Networking Conference, CCNC",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2025 IEEE 22nd Consumer Communications and Networking Conference, CCNC 2025",

}

TY - GEN

T1 - A Generalized XGBoost-Based Approach for Camera-Driven Beam Selection

AU - Zarei, Zahra

AU - Kang, Chung G.

PY - 2025

Y1 - 2025

N2 - Incorporating non-RF data in beam selection can speed up decision-making in situations that require a thorough search of all candidate options. Therefore, using camera sensory data at base stations to enhance millimeter-wave (mmWave) beam selection is becoming increasingly popular. This paper introduces a machine learning framework using a robust model along with XGBoost in beam prediction scenarios. Our approach enhances beam selection in both single and multi-candidate real-world vehicle-to-infrastructure (V2I) settings.

AB - Incorporating non-RF data in beam selection can speed up decision-making in situations that require a thorough search of all candidate options. Therefore, using camera sensory data at base stations to enhance millimeter-wave (mmWave) beam selection is becoming increasingly popular. This paper introduces a machine learning framework using a robust model along with XGBoost in beam prediction scenarios. Our approach enhances beam selection in both single and multi-candidate real-world vehicle-to-infrastructure (V2I) settings.

KW - Beam prediction

KW - Machine learning

KW - V2I

KW - XGBoost

UR - https://www.scopus.com/pages/publications/105005138250

U2 - 10.1109/CCNC54725.2025.10975896

DO - 10.1109/CCNC54725.2025.10975896

M3 - Conference contribution

AN - SCOPUS:105005138250

T3 - Proceedings - IEEE Consumer Communications and Networking Conference, CCNC

BT - 2025 IEEE 22nd Consumer Communications and Networking Conference, CCNC 2025

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 22nd IEEE Consumer Communications and Networking Conference, CCNC 2025

Y2 - 10 January 2025 through 13 January 2025

ER -

A Generalized XGBoost-Based Approach for Camera-Driven Beam Selection

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this