Optimizing Sparse RFI Prediction using Deep Learning

Abstract

Radio Frequency Interference (RFI) is an ever-present limiting factor among radio telescopes even in the most remote observing locations. When looking to retain the maximum amount of sensitivity and reduce contamination for Epoch of Reionization studies, the identification and removal of RFI is especially important. In addition to improved RFI identification, we must also take into account computational efficiency of the RFI-Identification algorithm as radio interferometer arrays such as the Hydrogen Epoch of Reionization Array grow larger in number of receivers. To address this, we present a Deep Fully Convolutional Neural Network (DFCN) that is comprehensive in its use of interferometric data, where both amplitude and phase information are used jointly for identifying RFI. We train the network using simulated HERA visibilities containing mock RFI, yielding a known "ground truth" dataset for evaluating the accuracy of various RFI algorithms. Evaluation of the DFCN model is performed on observations from the 67 dish build-out, HERA-67, and achieves a data throughput of 1.6$\times 10^5$ HERA time-ordered 1024 channeled visibilities per hour per GPU. We determine that relative to an amplitude only network including visibility phase adds important adjacent time-frequency context which increases discrimination between RFI and Non-RFI. The inclusion of phase when predicting achieves a Recall of 0.81, Precision of 0.58, and $F_2$ score of 0.75 as applied to our HERA-67 observations.

11 pages, 7 figures

References in corpus (17)

• Adam: A Method for Stochastic Optimization
• Fully Convolutional Networks for Semantic Segmentation
• Hydrogen Epoch of Reionization Array (HERA)
• GaLactic and Extragalactic All-sky Murchison Widefield Array (GLEAM) survey I: A low-frequency extragalactic catalogue
• A model of diffuse Galactic Radio Emission from 10 MHz to 100 GHz
• A morphological algorithm for improving radio-frequency interference detection
• A Per-Baseline, Delay-Spectrum Technique for Accessing the 21cm Cosmic Reionization Signature
• Fast Radio Burst 121102 Pulse Detection and Periodicity: A Machine Learning Approach
• Calibration Requirements for Detecting the 21 cm Epoch of Reionization Power Spectrum and Implications for the SKA
• The low-frequency environment of the Murchison Widefield Array: radio-frequency interference analysis and mitigation
• U-Net: Convolutional Networks for Biomedical Image Segmentation
• MITEoR: A Scalable Interferometer for Precision 21 cm Cosmology
• The LOFAR radio environment
• The impact of interference excision on 21-cm Epoch of Reionization power spectrum analyses
• A search for Fast Radio Bursts at low frequencies with Murchison Widefield Array high time resolution imaging
• Constraining Polarized Foregrounds for EoR Experiments I: 2D Power Spectra from the PAPER-32 Imaging Array
• Detecting Radio Frequency Interference in radio-antenna arrays with the Recurrent Neural Network algorithm

Cited by in corpus (26)

• Intelligent Computing: The Latest Advances, Challenges and Future
• Data Analysis for Precision 21 cm Cosmology
• Improved Constraints on the 21 cm EoR Power Spectrum and the X-Ray Heating of the IGM with HERA Phase I Observations
• Mitigating Internal Instrument Coupling II: A Method Demonstration with the Hydrogen Epoch of Reionization Array
• Absolving the SSINS of Precision Interferometric Radio Data: A New Technique for Mitigating Faint Radio Frequency Interference
• Redundant-Baseline Calibration of the Hydrogen Epoch of Reionization Array
• DAYENU: A Simple Filter of Smooth Foregrounds for Intensity Mapping Power Spectra
• Deep Learning improves identification of Radio Frequency Interference
• The Role of Machine Learning in the Next Decade of Cosmology
• Radio Frequency Interference Mitigation based on the ArPLS and SumThreshold Method
• Learning to detect RFI in radio astronomy without seeing it
• Quantifying Excess Power from Radio Frequency Interference in Epoch of Reionization Measurements
• Deep Learning Assisted Data Inspection for Radio Astronomy
• Measuring HERA's primary beam in-situ: methodology and first results
• Real-time RFI Filtering for uGMRT: Overview of the Released System and Relevance to the SKA
• Bayesian approach to radio frequency interference mitigation
• SKA-Low Intensity Mapping Pathfinder Updates: Deeper 21 cm Power Spectrum Limits from Improved Analysis Frameworks
• Characterization Of Inpaint Residuals In Interferometric Measurements of the Epoch Of Reionization
• Radio Frequency Interference Detection Using Efficient Multi-Scale Convolutional Attention UNet
• Performance of the Segment Anything Model in Various RFI/Events Detection in Radio Astronomy
• Temporal variability in low-frequency radio interference: Insight from high-cadence monitoring at a candidate radio notification zone in Malaysia
• TABASCAL: Removing multi-satellite interference from radio interferometry observations
• Noise Reduction Method for Radio Astronomy Single Station Observation Based on Wavelet Transform and Mathematical Morphology
• Imaging and Modeling Data from the Hydrogen Epoch of Reionization Array
• Operational Non-identifiability of Single-epoch Low-rank RFI Mitigation: Controlled Failure-mode Analysis and HERA Evidence
• Supervised Neural Networks for RFI Flagging