Javascript must be enabled for the correct page display

Proof of infinite causal horizon of the merging algorithms used in the XENONnT experiment

Dubrovskis, Stanislavs (2026) Proof of infinite causal horizon of the merging algorithms used in the XENONnT experiment. Bachelor's Thesis, Physics.

[img]
Preview
Text
StanislavsDubrovskisBScThesispublished.pdf

Download (1MB) | Preview

Abstract

Dark Matter makes up the majority of mass content of the universe yet due to its non-luminescent properties, its nature remains unknown. XENONnT experiment aims to detect the DM WIMPs or weakly interactive massive particles using a dual-phase xenon time projection chamber (TPC). Using the TPC, XENON collaboration wants to detect WIMP rare scattering events off a Xenon nuclei in the chamber. When such an event happens, XENONnT photomultiplier arrays detect two signal peaks: primary S1 and secondary S2. To reconstruct the S2 signal, a clustering algorithm merges several sub-peaks (peaklets) into one cluster. XENON researchers noticed that small perturbations in the data can have effect on the merging outcome at unexpectedly large distance from the perturbation origin. The thesis explores S2 peak merging algorithms used in the XENONnT detector’s data processing. In the thesis we study two generations of the merging algorithms and notice that both have the same hierarchical structure. This observation allows us to prove the existence of peaklet sequences (critical structures) for which local perturbations propagate arbitrarily far through merging, altering clustering decisions at arbitrarily large distances from the origin. Thus, the unbounded causal horizon of the algorithm is shown. We then test the XENONnT analysis pipeline with simulated critical structure and find it in the real data. Finally the consequences of this discovery are discussed and potential fixes are proposed.

Item Type: Thesis (Bachelor's Thesis)
Supervisor name: Aalbers, J. and Bruyn, K.A.M. de
Degree programme: Physics
Thesis type: Bachelor's Thesis
Language: English
Date Deposited: 11 Aug 2026 11:22
Last Modified: 11 Aug 2026 11:22
URI: https://fse.studenttheses.ub.rug.nl/id/eprint/37498

Actions (login required)

View Item View Item