CSI-Based Instantaneous Rainfall Monitoring in Sub-6 GHz for ISAC Applications

This dataset contains multi-antenna channel state information (CSI) measurements collected under natural rainfall events in the sub-6 GHz band. The dataset includes four temporally separated rainfall events (Case 1–Case 4) collected between August 11, 2025, and September 25, 2025, in Nanjing, China. Each CSI sample consists of a complex-valued matrix of dimension 3276 × 8, corresponding to 3276 OFDM subcarriers and 8 receive antennas. CSI samples were collected at 3-second intervals. Ground-truth rainfall intensity measurements were obtained using an OTT Parsivel² optical disdrometer with a temporal resolution of 10 seconds. A commercial 24 GHz Doppler weather radar dataset is also provided for comparison. To comply with the journal review process and ensure fair evaluation, the dataset was released in stages. In the current public release, Case 1 and Case 4 are available, while Case 2 and Case 3 are not included. How to Cite If you use this dataset in your research, please cite both the dataset and the associated publication. Dataset: Y. Li, J. Yang, W. Dong, C.-K. Wen, and S. Jin, “CSI-Based Instantaneous Rainfall Monitoring in Sub-6 GHz for ISAC Applications,” Zenodo, Dataset, 2026, doi: 10.5281/zenodo.18830281. Associated Publication: Y. Li, J. Yang, W. Dong, C.-K. Wen, and S. Jin, “CSI-Based Instantaneous Rainfall Monitoring in Sub-6 GHz for ISAC Applications,” IEEE Transactions on Communications, vol. 74, pp. 12418–12433, 2026, doi: 10.1109/TCOMM.2026.3720922. Related Publications For broader context on CSI-based rainfall sensing, including rainfall attenuation measurement and rainfall classification in sub-6 GHz systems, please refer to our earlier related work: Y. Li, J. Yang, Y. Huang, T. Yang, C.-K. Wen, and S. Jin, “RainGaugeNet: CSI-Based Sub-6 GHz Rainfall Attenuation Measurement and Classification for ISAC Applications,” IEEE Transactions on Communications, vol. 73, no. 12, pp. 13644–13658, 2025, doi: 10.1109/TCOMM.2025.3606633.