NTN-TDL-A-ModSet: A Modulation Recognition Dataset under 3GPP TR 38.811 NTN-TDL-A Channel

This dataset contains channel‑emulated recordings of 12 digital modulation schemes, generated under the **NTN-TDL-A** (Non‑Terrestrial Network Tapped‑Delay Line Type A) channel model defined in **3GPP TR 38.811**. It is specifically designed for **automatic modulation classification (AMC)** research under non‑terrestrial network (NTN) satellite communication scenarios, providing a realistic benchmark for evaluating deep learning models under NTN channel conditions.The data generation was performed by the **Software Defined System Studio (SDS Studio)** platform independently developed by the Broad‑band Communication and Network Group (BCNG) of the National University of Defense Technology, using a **channel emulator** configured according to 3GPP TR 38.811 specifications. All signals are generated under a **non‑line‑of‑sight (NLOS)** propagation condition, which is particularly relevant for satellite‑to‑ground communication links.The dataset is stored in a single HDF5 (`.h5`) or NumPy NPZ (`.npz`) file, with predefined training, validation, and test splits. Each sample consists of **I/Q complex samples** along with its modulation label and SNR value.