metadata
title: README
emoji: π
colorFrom: indigo
colorTo: gray
sdk: static
pinned: false
π RFInject: Synthetic RF Interference Injection for Sentinel-1 SAR L0 Data
π Dataset Links
| Resource | Type | Size | Link |
|---|---|---|---|
| RFInject Collection | Hugging Face Collection | β | ESA-philab/RFInject |
| RFInject-v1-L0 | Dataset | 395 GB | ESA-philab/RFInject-v1-L0 |
| RFInject-v1-SLC | Dataset | 23.8 TB | ESA-philab/RFInject-v1-SLC |
π Motivation
- Radio Frequency Interference (\gls{RFI}) is a major source of performance degradation in modern Synthetic Aperture Radar (\gls{SAR}) missions.
- The Copernicus Sentinel-1 constellation is significantly affected, with numerous studies reporting its detrimental impact.
- However, the lack of standardized and reproducible datasets has so far limited systematic benchmarking of RFI detection and mitigation strategies.
π οΈ What RFInject Brings
RFInject introduces a methodology for controlled synthetic RFI injection into clean Sentinel-1 L0 raw bursts, enabling:
- β Reproducible benchmarking of mitigation algorithms
- β Realistic simulation while retaining authentic system properties
- β Full parameter control over RFI characteristics
π Methodology Highlights
The framework is based on a parametric signal model:
- π― Synthetic RFI generation by superimposing modulated chirp trains onto authentic Sentinel-1 radar echoes.
- π§ Spectral and statistical fidelity ensured to reflect real operational systems.
- π Metadata-rich parameter sets controlling:
- π‘ Waveform diversity
- π Spatial extent
- β‘ Power scaling
π Dataset Features
- Clean Sentinel-1 L0 bursts β contaminated with controlled synthetic RFI
- Fully reproducible contamination scenarios
- Rich metadata for systematic testing across different algorithms and experimental setups
- Mirrored data access through the ESA Ξ¦-lab bucket storage infrastructure, with public entry point through the ESA Ξ¦-lab Hugging Face collection:
π https://huggingface.co/collections/ESA-philab/rfinject
π― Impact and Applications
The dataset empowers researchers to:
- π΅οΈββοΈ Detect RFI more reliably
- π‘οΈ Mitigate its impact effectively
- π€ Develop learning-based solutions for robust RFI-resilient SAR processing pipelines