Mitigating radio frequency interference in CHIME/FRB real-time intensity data

TL;DR

This paper introduces a novel iterative RFI mitigation technique for CHIME/FRB real-time data, significantly reducing interference effects and preserving data quality for radio transient surveys.

Contribution

The authors develop a new iterative method that effectively masks RFI in high-rate radio data, maintaining low false positive and negative rates, applicable to future radio transient surveys.

Findings

Reduces RFI-related bandwidth loss in CHIME/FRB data

Maintains low false positive and false negative rates

Applicable to high-performance radio transient surveys

Abstract

Extragalactic fast radio bursts (FRBs) are a new class of astrophysical transients with unknown origins that have become a main focus of radio observatories worldwide. FRBs are highly energetic ($\sim 10^{36}$-$10^{42}$ ergs) flashes that last for about a millisecond. Thanks to its broad bandwidth (400-800 MHz), large field of view ($\sim$200 sq. deg.), and massive data rate (1500 TB of coherently beamformed data per day), the Canadian Hydrogen Intensity Mapping Experiment / Fast Radio Burst (CHIME/FRB) project has increased the total number of discovered FRBs by over a factor 10 in 3 years of operation. CHIME/FRB observations are hampered by the constant exposure to radio frequency interference (RFI) from artificial devices (e.g., cellular phones, aircraft), resulting in $\sim$20% loss of bandwidth. In this work, we describe our novel technique for mitigating RFI in CHIME/FRB real-time intensity data. We mitigate RFI through a sequence of iterative operations, which mask out statistical outliers from frequency-channelized intensity data that have been effectively high-pass filtered. Keeping false positive and false negative rates at very low levels, our approach is useful for any high-performance surveys of radio transients in the future.