Real-time detection of Solar and Jovian radio bursts with NenuFAR: an overview of the TASKA-A use case in the EXTRACT project.

The EXTRACT Project has received funding from the European Union’s Horizon Europe programme under grant agreement number 101093110. Real-time detection of Solar and Jovian radio bursts with NenuFAR: advancing astrophysical data mining with the EXTRACT project. An overview of TASKA –A use cases Emilie Mauduit, Baptiste Cecconi, Cedric Viou, Louis Bondonneau, Julien Girard, Fadi Nammour TASKA –A1 : Solar spikes 3 Solar radio spikes •Dedicated to automatically detect solar radio spikes using SpikeNet [Murphy et al, OJA, 2024] SALF XI, Orléans –09/12/2025 4 Detection method : SpikeNet •ML model trained on ~100 000 samples of solar spikes (64 by 64 pixels) •Produced segmentation masks for radio spikes in the validation set and predicted the spike characteristics, i.e. location in time and frequency, duration, spectral width and drift rate •Bursts that are not fully in one tile can not be detected [Murphy et al, OJA, 2024] SALF XI, Orléans –09/12/2025 TASKA – A2 : Jovian S-bursts 6 •Developing an algorithm for real-time detection of fine drifting structures during radio observations with NenuFAR •Use a method originally developed for Jupiter observations with NDA-JunoN data [Mauduit et al, 2023, Nat Coms] The TASKA – A2 Use case SALF XI, Orléans –09/12/2025 7 1 - RFI mitigation 2 –FFT2D 3 - Radon transform Detection algorithm 8 MurMuRe pipeline and EXTRACT interfaces Conclusions 17 ➢Important step toward smart data filtering for next generation radiotelescopes such as NenuFAR or SKA. ➢It also paves the way for “analog to information” processing, which would drastically reduce the storage needs. ➢The type of emissions studied in this work require a high time-frequency resolution, but they are embedded within larger slowly-varying emissions that can be studied at a lower resolution. ➢This approach helps optimizing data storage while maintaining its value for scientific analysis, thus preparing for scalable solutions in the SKA era. ➢Code available on GitLab, easy to clone and use with notebook tutorials provided. Conclusions [https://gitlab.obspm.fr/extract] SALF XI, Orléans –09/12/2025 18 ➢Develop a CNN for S-bursts based on anomaly detection. ➢Optimize existing algorithms to have better detections ➢Find an even more efficient way to store the outputs. ➢A1 : include multi-scale processing Next steps and perspectives SALF XI, Orléans –09/12/2025 www.extract-project.eu www.extract-project.eu Follow us on social media: The EXTRACT Project has received funding from the European Union’s Horizon Europe programmeunder grant agreement number 101093110 Thank you ! [email protected]