SMART Framework: Real and Synthetic mmWave V2X Data for Beam Selection

Over 10,853 real-world samples and 26,600 synthetic samples for mmWave beam selection in vehicle-to-everything networks. The dataset, created by Muruganandham, Divyadharshini, consists of synchronized LiDAR, camera, and GPS data in diverse scenarios and a high-fidelity digital twin. It was last updated on 2025-10-15.

Use Cases

• Training mmWave beam selection models based on synchronized LiDAR, camera, and GPS data.
• Evaluating synthetic-to-real domain adaptation methods based on the paired e-FLASH and S-FLASH datasets.
• Developing meta-learning algorithms for V2X networks based on the diverse LOS and NLOS obstacle scenarios.
• Benchmarking ray-tracing simulations for wireless communication based on the Wireless InSite-generated synthetic data.

Strengths

• Includes over 10,853 real-world samples with synchronized multimodal data.
• Provides a high-fidelity synthetic counterpart with 26,600 samples generated using detailed ray-tracing.
• Covers diverse and relevant mmWave V2X scenarios including LOS and NLOS with various obstacles.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count for the combined dataset is unknown, which may limit suitability assessment.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Texas Data Repository Harvested Dataverse

Collection Method

Real data captured in diverse mmWave V2X scenarios; synthetic data generated using Blender, Blensor, and Wireless InSite.

Time Range

null

Freshness

Last updated 2025-10-15 03:40:11; freshness should be verified.

Geography

null