We report on a search for engineered signals from a sample of 692 nearby stars using the Robert C. Byrd Green Bank Telescope, undertaken as part of the Breakthrough Listen Initiative search for extraterrestrial intelligence. Observations were made over 1.1-1.9 GHz (L band), with three sets of five-minute observations of the 692 primary targets, interspersed with five-minute observations of secondary targets. By comparing the "ON" and "OFF" observations, we are able to identify terrestrial interference and place limits on the presence of engineered signals from putative extraterrestrial civilizations inhabiting the environs of the target stars. During the analysis, 11 events passed our thresholding algorithm, but a detailed analysis of their properties indicates that they are consistent with known examples of anthropogenic radio-frequency interference. We conclude that, at the time of our observations, none of the observed systems host high-duty-cycle radio transmitters emitting between 1.1 and 1.9 GHz with an Equivalent Isotropic Radiated Power of ∼10 13 W, which is readily achievable by our own civilization. Our results suggest that fewer than ∼0.1% of the stellar systems within 50 pc possess the type of transmitters searched in this survey.

Breakthrough Listen (BL) is a ten-year initiative to search for signatures of technologically capable life beyond Earth via radio and optical observations of the local Universe. A core part of the BL program is a comprehensive survey of 1702 nearby stars at radio wavelengths (1-10 GHz). Here, we report on observations with the 64-m CSIRO Parkes radio telescope in New South Wales, Australia, and the 100-m Robert C. Byrd Green Bank radio telescope in West Virginia, USA. Over 2016 January to 2019 March, a sample of 1138 stars was observed at Green Bank using the 1.10-1.90 GHz and 1.80-2.80 GHz receivers, and 189 stars were observed with Parkes over 2.60-3.45 GHz. We searched these data for the presence of engineered signals with Doppler-acceleration drift rates between ±4 Hz s −1 . Here, we detail our data analysis techniques and provide examples of detected events. After excluding events with characteristics consistent with terrestrial radio interference, we are left with zero candidates. That is, we find no evidence of putative radio transmitters above 2.1×10 12 W, and 9.1×10 12 W for Green Bank and Parkes observations, respectively. These observations constitute the most comprehensive search over 1.10-3.45 GHz for technosignatures to date. All data products, totalling ∼219 TB, are available for download as part of the first BL data release (DR1), as described in a companion paper (Lebofsky et. al., 2019)

A radio transmitter which is accelerating with a non-zero radial component with respect to a receiver will produce a signal that appears to change its frequency over time. This effect, commonly produced in astrophysical situations where orbital and rotational motions are ubiquitous, is called a drift rate. In radio SETI (Search for Extraterrestrial Intelligence) research, it is unknown a priori which frequency a signal is being sent at, or even if there will be any drift rate at all besides motions within the solar system. Therefore a range of potential drift rates need to be individually searched, and a maximum drift rate needs to be chosen. The middle of this range is zero, indicating no acceleration, but the absolute value for the limits remains unconstrained. A balance must be struck between computational time and the possibility of excluding a signal from an ETI. In this work, we examine physical considerations that constrain a maximum drift rate and highlight the importance of this problem in any narrowband SETI search. We determine that a normalized drift rate of 200 nHz (eg. 200 Hz/s at 1 GHz) is a generous, physically motivated guideline for the maximum drift rate that should be applied to future narrowband SETI projects if computational capabilities permit.