Radio frequency interference mitigation using deep convolutional neural networks

• Published in: Astronomy and computing Vol. 18; pp. 35 - 39
• Main Authors: Akeret, J., Chang, C., Lucchi, A., Refregier, A.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.01.2017
• Subjects: Convolutional neural network; Deep learning; Radio frequency interference; RFI mitigation; Radio frequency interference; RFI mitigation; Deep learning; Convolutional neural network
• ISSN: 2213-1337; 2213-1345
• DOI: 10.1016/j.ascom.2017.01.002

We propose a novel approach for mitigating radio frequency interference (RFI) signals in radio data using the latest advances in deep learning. We employ a special type of Convolutional Neural Network, the U-Net, that enables the classification of clean signal and RFI signatures in 2D time-ordered data acquired from a radio telescope. We train and assess the performance of this network using the HIDE &SEEK radio data simulation and processing packages, as well as early Science Verification data acquired with the 7m single-dish telescope at the Bleien Observatory. We find that our U-Net implementation is showing competitive accuracy to classical RFI mitigation algorithms such as SEEK’s SumThreshold implementation. We publish our U-Net software package on GitHub under GPLv3 license.