Searching for broadband pulsed beacons from 1883 stars using neural networks

TL;DR

This paper reports a large-scale search for broadband pulsed beacons from 1883 stars using neural networks, demonstrating the method's effectiveness and placing constraints on the prevalence of such signals in our neighborhood.

Contribution

It introduces the first extensive survey for broadband pulsed extraterrestrial signals utilizing neural networks and GPU acceleration, expanding the search methods in SETI.

Findings

No signals of interest detected.

Constraints on the existence of broadband pulsed beacons in the local neighborhood.

Neural network classifier effectively processed large-scale radio data.

Abstract

The search for extraterrestrial intelligence at radio frequencies has largely been focused on continuous-wave narrowband signals. We demonstrate that broadband pulsed beacons are energetically efficient compared to narrowband beacons over longer operational timescales. Here, we report the first extensive survey searching for such broadband pulsed beacons towards 1883 stars as a part of the Breakthrough Listen's search for advanced intelligent life. We conducted 233 hours of deep observations across 4 to 8 GHz using the Robert C. Byrd Green Bank Telescope and searched for three different classes of signals with artificial (or negative) dispersion. We report a detailed search -- leveraging a convolutional neural network classifier on high-performance GPUs -- deployed for the very first time in a large-scale search for signals from extraterrestrial intelligence. Due to the absence of any signal-of-interest from our survey, we place a constraint on the existence of broadband pulsed beacons in our solar neighborhood: $\lesssim$1 in 1000 stars have transmitter power-densities $\gtrsim$10$^5$ W/Hz repeating $\leq$500 seconds at these frequencies.