Radio Frequency Source Localization Data via Learned Path-Loss Coefficients

A data-driven framework learns path-loss coefficients from real-world received signal power measurements and terrain features from satellite imagery. The 20.1 KB dataset, authored by Karunamuni Sahanya Udani Silva and last updated in May 2026, supports a 3D multilateration method for locating RF transmitters. Experimental results indicate improved localization accuracy and generalization over traditional models.

Use Cases

• Training machine learning models to predict path-loss coefficients based on signal power measurements and terrain features.
• Evaluating 3D multilateration algorithms for RF transmitter localization using learned distance estimates.
• Benchmarking the accuracy of data-driven path-loss models against traditional empirical models.
• Analyzing the relationship between satellite-derived terrain features and radio signal propagation characteristics.

Strengths

• Dataset is small (20.1 KB), facilitating quick download and initial exploration.
• Methodology is described as using real-world measurements and publicly available satellite imagery.
• License is CC-BY-4.0, permitting broad sharing and adaptation.

Limitations

• Row count and specific column definitions are unknown, limiting suitability assessment.
• Description metadata is limited; actual data quality and structure require manual inspection after download.
• The dataset's spatial and temporal coverage for the signal measurements is not specified.

Provenance

Source

figshare

Collection Method

Data-driven framework using real-world received signal power measurements and terrain features from satellite imagery.

Freshness

Last updated 2026-05-28 13:58:33