metadata
title: README
emoji: 🌍
colorFrom: indigo
colorTo: gray
sdk: static
pinned: false
🚀 RFInject: Synthetic RF Interference Injection for Sentinel-1 SAR L0 Data
📌 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.
🛠️ What RFInject Brings
RFInject introduces a methodology for controlled synthetic RFI injection into clean Sentinel-1 L0 raw bursts, enabling:
- ✅ Reproducible benchmarking of mitigation algorithms
- ✅ Realistic simulation while retaining authentic system properties
- ✅ Full parameter control over RFI characteristics
📐 Methodology Highlights
The framework is based on a parametric signal model:
- 🎯 Synthetic RFI generation by superimposing modulated chirp trains onto authentic Sentinel-1 radar echoes.
- 🧠 Spectral and statistical fidelity ensured to reflect real operational systems.
- 📊 Metadata-rich parameter sets controlling:
- 📡 Waveform diversity
- 🌍 Spatial extent
- ⚡ Power scaling
📂 Dataset Features
- Clean Sentinel-1 L0 bursts → contaminated with controlled synthetic RFI
- Fully reproducible contamination scenarios
- Rich metadata for systematic testing across different algorithms and experimental setups
🎯 Impact and Applications
The dataset empowers researchers to:
- 🕵️♂️ Detect RFI more reliably
- 🛡️ Mitigate its impact effectively
- 🤖 Develop learning-based solutions for robust RFI-resilient SAR processing pipelines