arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2603.15093eess.SPcs.ITmath.IT
收藏

基于多模态大语言模型的波束预测

Beam Prediction Based on Multimodal Large Language Models

Southeast University(东南大学);Purple Mountain Laboratories(紫金山实验室);Imperial College London(帝国理工学院)

Tianhao Mao, Le Liang, Jie Yang, Xiao Li, Shi Jin, Geoffrey Ye Li

AI总结:

本文提出基于多模态大语言模型的波束预测框架,通过融合RGB图像和LiDAR点云等异构数据,提升动态环境下波束预测精度与通信性能,实验表明其在波束准确性和通信性能上优于现有方法。

摘要:

准确的波束预测是下一代无线通信系统的关键使能技术。本文提出了一种基于多模态大语言模型(LLM)的波束预测框架,有效利用由RGB相机图像和LiDAR点云等传感器数据提供的上下文信息。为有效融合异构模态,我们设计了专用模态编码器,结合波束引导的注意力掩码机制和高频时间对齐策略,实现动态环境下的稳健跨模态特征集成。此外,我们构建了一个大规模多模态通信数据集,名为Multimodal-Wireless,涵盖多样化的天气和交通条件,并具有高保真的射线追踪标签。广泛仿真结果表明,所提出的方法显著减少了对oracle角度出发知识的依赖,并在波束准确性和通信性能方面优于最先进的多模态LLM波束预测方法,将平均Top-1准确率提升到80.8%,平均归一化增益提升到89.1%。

英文摘要:

Accurate beam prediction is a key enabler for next-generation wireless communication systems. In this paper, we propose a multimodal large language model (LLM)-based beam prediction framework that effectively utilizes contextual information, provided by sensory data including RGB camera images and LiDAR point clouds. To effectively fuse heterogeneous modalities, we design specialized modality encoders together with a beam-guided attention masking mechanism and a high-frequency temporal alignment strategy, enabling robust cross-modal feature integration under dynamic environments. Furthermore, we construct a large-scale multimodal dataset for communication, named Multimodal-Wireless, which covers diverse weather and traffic conditions with high-fidelity ray-tracing labels. Extensive simulation results demonstrate that the proposed approach significantly reduces the reliance on oracle angle-of-departure knowledge and consistently outperforms state-of-the-art multimodal LLM-based beam prediction methods in terms of beam accuracy and communication performance, improving the average Top-1 accuracy to 80.8% and the average normalized gain to 89.1%.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑