WiWorld-RealData: A Real-World Multi-Modal Dataset for 6G Wireless World Models

Abstract

Wireless world models aim to represent, predict, and reason about wireless propagation by jointly understanding physical environments and channel responses. Realizing such models in sixth-generation (6G) digital twin channels requires datasets that capture measured wireless responses and environment states under real-world propagation conditions. This paper presents WiWorld-RealData, a real-world outdoor multi-band channel and multi-modal sensing dataset collected along campus mobile routes. WiWorld-RealData provides measured channel impulse responses (CIRs) at 3.7 GHz and 6.775 GHz, together with multi-view images, panoramic images, light detection and ranging (LiDAR) point clouds, millimeter-wave (mmWave) radar records, and global navigation satellite system (GNSS) trajectories. Through unified file organization and metadata manifests, the dataset establishes sample-level correspondences among channel responses, environment observations, timestamps, route information, antenna configurations, and quality flags. The overall measurement campaign has produced 10 TB-level multi-modal field data. The current public release provides one representative dual-band route at 3.7 GHz and 6.775 GHz with complete channel-environment alignment, while the acquisition framework supports extension to more frequency bands and scenarios. A case study on environment-assisted path-loss prediction achieves a mean absolute error (MAE) of 2.02 dB and a root mean squared error (RMSE) of 2.69 dB, indicating that the aligned environment observations contain predictive information for channel variations. The dataset is available at https://scc.bupt.edu.cn/dataset-manage/datasets/44, and a ScienceDB mirror will be provided upon release.