CorteXforge: Scenario-Based RF Dataset Generation on Real Testbeds
Abstract
Deep learning for radio frequency (RF) signal analysis relies on large annotated datasets, yet synthetic datasets trade physical realism for controllability, while uncontrolled over-theair measurements offer realism but limited reproducibility. We present CorteXforge, a scenario-based framework for generating realistic and reproducible RF datasets on programmable software defined radio (SDR) testbeds. High-level RF scenarios are automatically translated into synchronized experiments executed on the SLICES-RI/CorteXlab platform, producing fully annotated Signal Metadata Format (SIGMF) datasets that capture hardware impairments and propagation effects. We validate CorteXforge through automatic modulation classification using ModForge, an over-the-air (OTA) dataset generated with the framework. Cross-domain experiments reveal a synthetic-to-real performance gap and show that synthetic pretraining remains beneficial for adaptation to OTA data. CorteXforge provides a practical foundation for reproducible RF dataset generation and future research on domain generalization.
Related
- Fused ultrametric Gromov-Wasserstein for Scenario Tree generation in Multistage Stochastic Programming
- Feasible closed form moment matching for scenario generation through covariance rotations
- ISOMORPH: A Supply Chain Digital Twin for Simulation, Dataset Generation, and Forecasting Benchmarks
- Should I Use This Synthetic Dataset for Training? How to Test with Minimal Real Data
- Integrability in Asymptotic Symmetries of Spacetime: the BMS3 scenario
BibTeX
@misc{joly2026cortexforge,
title = {{CorteXforge: Scenario-Based RF Dataset Generation on Real Testbeds}},
author = {Andréa Joly and Leonardo S Cardoso and Jean-Marie Gorce},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.26999},
eprint = {2610.26999},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.26999},
}
RIS
TY - UNPB
TI - CorteXforge: Scenario-Based RF Dataset Generation on Real Testbeds
AU - Andréa Joly
AU - Leonardo S Cardoso
AU - Jean-Marie Gorce
PY - 2026
DA - 2026/10/19
PB - arXiv
DO - 10.48550/arxiv.2610.26999
UR - https://arxiv.org/abs/2610.26999
AB - Deep learning for radio frequency (RF) signal analysis relies on large annotated datasets, yet synthetic datasets trade physical realism for controllability, while uncontrolled over-theair measurements offer realism but limited reproducibility. We present CorteXforge, a scenario-based framework for generating realistic and reproducible RF datasets on programmable software defined radio (SDR) testbeds. High-level RF scenarios are automatically translated into synchronized experiments executed on the SLICES-RI/CorteXlab platform, producing fully annotated Signal Metadata Format (SIGMF) datasets that capture hardware impairments and propagation effects. We validate CorteXforge through automatic modulation classification using ModForge, an over-the-air (OTA) dataset generated with the framework. Cross-domain experiments reveal a synthetic-to-real performance gap and show that synthetic pretraining remains beneficial for adaptation to OTA data. CorteXforge provides a practical foundation for reproducible RF dataset generation and future research on domain generalization.
ER -