Has created a new machine learning tool UC Berkeley And the SETI Institute mimics what an extraterrestrial intelligence message received on Earth would look like.

It is known as Setigen and has a library Open source Which could change the rules of the game for future exploration of intelligence outside of Earth (abbreviated SETI in English).

Their purpose has been since 1960 Research Detection of non-human technical signatures, such as emissions produced by technology in radio observation.

Although many have speculated about such a character Technology companiesRadio-SETI basically involves searching for signals that have never been detected, but have a wide range of possible sizes.

Since we have not yet identified a SETI radio signal, we must estimate its size in order to develop search algorithms. Lack Positive identification It also makes it difficult to test the inherent efficiency of these algorithms.

To address these challenges, Setigen has been introduced as an open source Python-based library for heuristic-based signal synthesis and injection. Spectrogram (Dynamic Spectra) and raw voltage data, according to the new program submission to arXiv.

“What Setigen does is facilitate production Synthetic SETI signalWhich can be used on fully synthetic data or combined with real observational data to provide more realistic RFI background noise, “explains Brian Brozicki, a graduate student. Astronomy Berkeley and the team leader developing the initiative.

“In this way, we can create large datasets of synthetic signals to analyze the sensitivity of existing algorithms or to serve as the basis for it. Training Machine learning, “he added in a statement to Universe Today.


By admin

Leave a Reply

Your email address will not be published.