Wireless World Model - Test Dataset

Dataset Overview This dataset provides testing data for the study “Wireless World Models for Future AI-Native Networks”. It contains simulated channel state information (CSI) matrices sampled across multiple time steps, enabling the evaluation of learning-based wireless channel modeling and prediction methods. The dataset corresponds to the experiments reported in the associated paper. The testing data are designed to evaluate model performance under three conditions: in-distribution testing, velocity generalization, and urban scenario generalization. The dataset includes three subsets: test_regular: in-distribution testing data generated under the same environments and velocity settings as the training data. test_gen_velocity: velocity generalization testing data with UE velocities of 40 km/h and 70 km/h, which are not present in the training dataset. test_gen_city: scenario generalization testing data generated in a new urban environment, Wall Street, New York, which is not included in the training environments. Similar to the training dataset, channel realizations are densely sampled over time and synchronized with the spatial coordinates of user equipment (UE), which are provided in the accompanying position (pos) files. This design yields time-continuous CSI–position pairs rather than independent channel snapshots. Data Generation Method The CSI data were generated using the Sionna ray-tracing framework, which provides physically consistent wireless channel simulation. The ray-tracing environment models realistic urban propagation conditions including reflections, diffractions, and multi-path propagation. File Organization dataset/ ├── test_regular.zip │ ├── H/ │ │ ├── samples_H_... │ └── pos/ │ └──samples_pos_... ├── test_gen_velocity.zip │ ├── H/ │ └── pos/ └── test_gen_city.zip ├── H/ └── pos/ test_regular.zip contains in-distribution testing samples. test_gen_velocity.zip contains testing samples with unseen UE velocities (40 km/h and 70 km/h). test_gen_city.zip contains testing samples generated in the unseen urban environment of Wall Street, New York. Data Format Each dataset file is stored in NumPy (.npy) format. The CSI data file samples_H_xxx_bsx_xxkmh_2048_seedxxxxxxxxxx.npy has the following shape: (2048, 2, 16, 32, 32) where 2048 : number of channel samples2 : real and imaginary components16 : temporal snapshots32 : base station antennas features32 : user equipment antennas and frequency subbands features The real and imaginary parts are stored separately along the second dimension. The position data file samples_pos_xxx_bsx_xxkmh_2048_seedxxxxxxxxxx.npy has the shape: (2048,) The file contains 2048 dictionary entries, where each dictionary corresponds to one CSI sample. Each dictionary includes the following keys: ue_traj: the trajectory of the user equipment (UE), storing the UE positions at 16 time steps. cityname: the name of the urban scenario in which the data are generated. bs_pos: the spatial coordinates of the base station used in the simulation. Usage Unzip the dataset files into the directory src/datasets/raw_data/ of the wireless_world_model_v1 code repository. Then run the script preprocess_example_script.py and set RAW_DATASET_NAME='test_<type>' to perform dataset preprocessing, where <type> corresponds to regular, gen_velocity, or gen_city.