A search for technosignatures from TRAPPIST-1, LHS 1140, and 10 planetary systems in the Kepler field with the Green Bank Telescope at 1.15-1.73 GHz

TL;DR

This study conducted a comprehensive search for technosignatures in multiple planetary systems using the Green Bank Telescope, developing an improved detection algorithm that significantly increased candidate detections and addressed common analysis issues.

Contribution

The paper introduces an enhanced data processing algorithm that improves detection efficiency and addresses challenges in radio technosignature searches, applied to data from the Green Bank Telescope.

Findings

Over 98% of signals identified as RFI

30 terrestrial-origin candidates detected outside RFI regions

Algorithm increased candidate detections by over four times

Abstract

As part of our ongoing search for technosignatures, we collected over three terabytes of data in May 2017 with the L-band receiver (1.15-1.73 GHz) of the 100 m diameter Green Bank Telescope. These observations focused primarily on planetary systems in the Kepler field, but also included scans of the recently discovered TRAPPIST-1 and LHS 1140 systems. We present the results of our search for narrowband signals in this data set with techniques that are generally similar to those described by Margot et al. (2018). Our improved data processing pipeline classified over $98\%$ of the $\sim$ 6 million detected signals as anthropogenic Radio Frequency Interference (RFI). Of the remaining candidates, 30 were detected outside of densely populated frequency regions attributable to RFI. These candidates were carefully examined and determined to be of terrestrial origin. We discuss the problems associated with the common practice of ignoring frequency space around candidate detections in radio technosignature detection pipelines. These problems include inaccurate estimates of figures of merit and unreliable upper limits on the prevalence of technosignatures. We present an algorithm that mitigates these problems and improves the efficiency of the search. Specifically, our new algorithm increases the number of candidate detections by a factor of more than four compared to Margot et al. (2018).