Saved in:
Bibliographic Details
Main Authors: Ma, Mengyuan, Welgamage, Isuri, Alkhateeb, Ahmed, Swindlehurst, A. Lee, Juntti, Markku, Nguyen, Nhan Thanh
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.16708
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913042000773120
author Ma, Mengyuan
Welgamage, Isuri
Alkhateeb, Ahmed
Swindlehurst, A. Lee
Juntti, Markku
Nguyen, Nhan Thanh
author_facet Ma, Mengyuan
Welgamage, Isuri
Alkhateeb, Ahmed
Swindlehurst, A. Lee
Juntti, Markku
Nguyen, Nhan Thanh
contents Beam training and prediction in real-world millimeter-wave (mmWave) communications systems are challenging due to rapidly time-varying channels and strong interference from surrounding objects. In this context, widely available sensors, such as cameras and radars, can capture rich environmental information, enabling efficient beam management. This paper proposes a knowledge-distillation (KD)-enabled learning framework for developing lightweight and low-complexity models for beam prediction and tracking using real-world camera and radar data from the DeepSense 6G dataset. Specifically, a powerful teacher network based on convolutional neural networks (CNNs) and gated recurrent units (GRUs) is first designed to predict current and future beams from historical sensor observations. Then, a compact student model is constructed and trained via KD to transfer the predictive capability of the teacher model to a lightweight architecture. Simulation results demonstrate that jointly leveraging radar and image modalities significantly outperforms single-modality approaches. Moreover, the proposed student model achieves over 96% Top-5 beam prediction accuracy while reducing computational complexity by more than 4 times and the number of parameters by over 27 times compared with the teacher model.
format Preprint
id arxiv_https___arxiv_org_abs_2604_16708
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Knowledge Distillation for Lightweight Multimodal Sensing-Aided mmWave Beam Tracking
Ma, Mengyuan
Welgamage, Isuri
Alkhateeb, Ahmed
Swindlehurst, A. Lee
Juntti, Markku
Nguyen, Nhan Thanh
Signal Processing
Beam training and prediction in real-world millimeter-wave (mmWave) communications systems are challenging due to rapidly time-varying channels and strong interference from surrounding objects. In this context, widely available sensors, such as cameras and radars, can capture rich environmental information, enabling efficient beam management. This paper proposes a knowledge-distillation (KD)-enabled learning framework for developing lightweight and low-complexity models for beam prediction and tracking using real-world camera and radar data from the DeepSense 6G dataset. Specifically, a powerful teacher network based on convolutional neural networks (CNNs) and gated recurrent units (GRUs) is first designed to predict current and future beams from historical sensor observations. Then, a compact student model is constructed and trained via KD to transfer the predictive capability of the teacher model to a lightweight architecture. Simulation results demonstrate that jointly leveraging radar and image modalities significantly outperforms single-modality approaches. Moreover, the proposed student model achieves over 96% Top-5 beam prediction accuracy while reducing computational complexity by more than 4 times and the number of parameters by over 27 times compared with the teacher model.
title Knowledge Distillation for Lightweight Multimodal Sensing-Aided mmWave Beam Tracking
topic Signal Processing
url https://arxiv.org/abs/2604.16708