Artificial intelligence (AI) and deep learning have made a full impact on society the last decades, including the realm of particle physics. This thesis explores whether a neural network, a deep learning program mimicking the biological brain, can be used to reject noise in real time at the Radio Neutrino Observatory in Greenland (RNO-G). RNO-G aims to detect radio waves in the ice cape of Greenland, induced by ultra high energy neutrinos ($>10^{18}$ eV). Due to the low flux of neutrinos at these energies, it is desired to increase the sensititivty of RNO-G by lowering the trigger threshold as much as possible. However, lowering the threshold is currently limited by unavoidable thermal noise fluctuations that would otherwise saturate the detector. Previous research has shown that a neural network could be used on a similar neutrino detector, ARIANNA, to reject thermal noise in real time, thus making it possible to lower the trigger threshold below the noise floor. This thesis aims to do the same for RNO-G.
Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:uu-480020 |
Date | January 2022 |
Creators | Liland, Lukas |
Publisher | Uppsala universitet, Institutionen för fysik och astronomi |
Source Sets | DiVA Archive at Upsalla University |
Language | English |
Detected Language | English |
Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
Format | application/pdf |
Rights | info:eu-repo/semantics/openAccess |
Relation | FYSAST ; FYSKAND1150 |
Page generated in 0.002 seconds