Multi-Modal Sensing-Aided mmWave Beam Tracking via Deep Reinforcement Learning for 6G
IEEE Communications Letters · 2026 · 30 · 2805-2809
DOI: 10.1109/lcomm.2026.3719400 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Communications Letters, 2026. Verfasst von Yue Su und 3 weiteren Autor:innen. Die Arbeit nennt 14 Literaturangaben. Gefördert von: Program for Innovative Research Team in Universities of Inner Mongolia Autonomous Region, Natural Science Foundation of Inner Mongolia Autonomous Region, High-Level Talents Recruitment Project of Inner Mongolia University.
Abstract (Originalsprache)
Multi-modal heterogeneous information, such as sub-6 GHz channels, millimeter wave (mmWave) channels, and 3D point clouds, can guide communication system optimization by providing real-time feedback on changes in the physical environment. This letter proposes a multi-modal sensing-assisted mmWave beam-tracking (MMS-BT) method based on deep reinforcement learning. By leveraging multi-modal sensing information, the beam-tracking range can be dynamically adjusted, thereby enabling environment-adaptive beam-tracking decisions in mmWave channels with spatially consistent properties. Simulation results show that multi-modal sensing enhances the robustness of AI-assisted beam tracking. Furthermore, fusing 3D point clouds with radio-frequency sensing can improve the beam-tracking success probability by 13.1%.
Förderung
- Program for Innovative Research Team in Universities of Inner Mongolia Autonomous Region
- Natural Science Foundation of Inner Mongolia Autonomous Region
- High-Level Talents Recruitment Project of Inner Mongolia University
Zitieren
APA
Su, Y., Zhang, Z., Yu, L., & Cai, W. (2026). Multi-Modal Sensing-Aided mmWave Beam Tracking via Deep Reinforcement Learning for 6G. IEEE Communications Letters, 30, 2805-2809. https://doi.org/10.1109/lcomm.2026.3719400
Vancouver
Su Y, Zhang Z, Yu L, Cai W. Multi-Modal Sensing-Aided mmWave Beam Tracking via Deep Reinforcement Learning for 6G. IEEE Communications Letters. 2026;30:2805-2809. doi:10.1109/lcomm.2026.3719400
BibTeX
@article{Su2026,
author = {Su, Yue and Zhang, Zhen and Yu, Li and Cai, Wenhao},
title = {Multi-Modal Sensing-Aided mmWave Beam Tracking via Deep Reinforcement Learning for 6G},
journal = {IEEE Communications Letters},
year = {2026},
volume = {30},
pages = {2805--2809},
doi = {10.1109/lcomm.2026.3719400}
}
EUVIMED