Vishal Gajjar
September 29, 2022
Abstract:
For the past six decades, radio SETI has largely been focused on continuous-wave narrowband signals. In Gajjar et al. (2022), we demonstrated that broadband pulsed beacons are energetically efficient compared to narrowband beacons over longer operational timescales. We also reported the first extensive survey searching for such broadband pulsed beacons towards close to 2000 stars with the GBT. In this talk, I will talk about our detailed search strategy leveraging a convolutional neural network classifier on high-performance GPUs deployed for the very first time in a large-scale search for signals from ETIs. Using this classifier, we were able to reduce the number of false positives by 97%. Our selection criteria did not return any signals of interest. We thus place a constraint on the existence of broad-band pulsed beacons in our solar neighbourhood: <1 in 1000 stars have transmitter power densities >10^5 W/Hz.
For information about the speaker’s other work: gajjarvishal.com