FARM: Foundational Aerial Radio Map for Low-Altitude Networking

FARM Aerial Radio Map (ARM) Dataset is a collection of radio propagation maps for intelligent low-altitude networking. The dataset, released by author jliang097, is based on the ARM-Omni model and includes data for training, in-domain evaluation (D1-D10), and zero-shot evaluation (P1, F1, A1). It was last updated on Hugging Face on 2026-05-04.

Use Cases

• Training foundational models for radio signal prediction based on the ARM-Omni framework.
• Evaluating model performance on in-domain scenarios using the D1-D10 evaluation subsets.
• Conducting zero-shot generalization tests using the P1, F1, and A1 evaluation subsets.
• Analyzing radio wave propagation at different frequencies and beamwidths for low-altitude networks.

Strengths

• Dataset is explicitly designed for foundational model training and evaluation, as described in the associated arXiv paper.
• Includes distinct subsets for training, in-domain evaluation, and zero-shot evaluation, suggesting a structured benchmarking approach.
• Last updated on 2026-05-04, indicating recent maintenance.

Limitations

• Description metadata is limited; actual data quality, file formats, and size require manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and specific geographic coverage are unknown, which may limit suitability assessment.

Provenance

Source

jliang097 on Hugging Face.

Collection Method

Constructed based on the ARM-Omni model, as per the associated research paper.

Freshness

Last updated 2026-05-04 18:26:39; freshness should be verified.