Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Pattern-Based Electron Counting Algorithm for LDMX

Kvarnstrom Meissner, Lucia LU (2026) FYSK04 20261
Department of Physics
Particle and nuclear physics
Abstract
The Trigger Scintillator (TS) of the Light Dark Matter eXperiment (LDMX) is responsible for counting the electrons in the incoming beam in order to accurately determine the trigger threshold used in identifying missing momentum events. Electrons are counted by reconstructing their approximately horizontal tracks through the TS. The software algorithm operates on a maximum vertical tolerance value, making it both computationally expensive and inflexible to a misalignment in the TS geometry.

This thesis proposes an alternate counting algorithm using a data-driven lookup table (LUT) containing real electron patterns derived from simulation or data. This allows the tracking definition to naturally adapt to the TS geometry presented in the... (More)
The Trigger Scintillator (TS) of the Light Dark Matter eXperiment (LDMX) is responsible for counting the electrons in the incoming beam in order to accurately determine the trigger threshold used in identifying missing momentum events. Electrons are counted by reconstructing their approximately horizontal tracks through the TS. The software algorithm operates on a maximum vertical tolerance value, making it both computationally expensive and inflexible to a misalignment in the TS geometry.

This thesis proposes an alternate counting algorithm using a data-driven lookup table (LUT) containing real electron patterns derived from simulation or data. This allows the tracking definition to naturally adapt to the TS geometry presented in the data. To write the LUT, electrons’ characteristic propagation patterns are first examined using truth-level information and then compared to unfiltered simulation data. It is shown that real tracks can be isolated from random combinatorics patterns by applying a minimum pattern frequency threshold to simulated data. At an optimal threshold of 0.0008, the data-driven LUT is able to achieve a maximum tracking efficiency 99.37% for a fake track rate 0.001668%, compared to the tolerance definition algorithm’s efficiency 98.00% and fake rate 0.00073%. This study demonstrates the feasibility of a data driven tracking method, with potential for further improvement through refinement of the frequency-threshold definition. (Less)
Popular Abstract
The mysterious dark matter has puzzled scientists ever since its initial discovery. The story began with observing galaxy behavior, and today, scientists believe that this invisible mass is composed of particles, whose natures remain largely unknown. Accelerator-based searches for dark particles often turn up empty-handed, but the Light Dark Matter eXperiment has proposed to extend the dark matter search to a new mass range, using an electron beam shot toward a tungsten target to observe the
production of dark particles. This process will be indicated by a large loss of momentum, but it is extremely rare. To filter the rare events of interest out of the otherwise enormous data pool, LDMX uses a trigger system, including a trigger... (More)
The mysterious dark matter has puzzled scientists ever since its initial discovery. The story began with observing galaxy behavior, and today, scientists believe that this invisible mass is composed of particles, whose natures remain largely unknown. Accelerator-based searches for dark particles often turn up empty-handed, but the Light Dark Matter eXperiment has proposed to extend the dark matter search to a new mass range, using an electron beam shot toward a tungsten target to observe the
production of dark particles. This process will be indicated by a large loss of momentum, but it is extremely rare. To filter the rare events of interest out of the otherwise enormous data pool, LDMX uses a trigger system, including a trigger scintillator (TS). The job of the trigger is to set a minimum energy threshold, and only those events over that threshold are saved for later analysis. In order to set this threshold, the number of incoming electrons must be known. It is the TS which is therefore responsible for electron counting.

Electrons are counted by reconstructing their tracks left behind in the three TS modules. One track tallies as one electron. The current track-making algorithm relies on the assumption that these modules are, ideally, perfectly aligned. Under realistic detector conditions, this is unlikely. This thesis introduces a new track-making algorithm which stores real, previously observed track patterns in a lookup table (LUT), allowing any detector data to adapt to a potential misalignment. The LUT acts as an answer sheet containing the real track patterns, and tracks can then be made by comparing combinations of detected signals with the combinations in the LUT—if they exist, they are accepted as tracks, and if not, they are rejected. Noise may originate from background processes that interfere with track making, therefore this project begins with using true electron track patterns. They are confirmed to be true patterns by the information provided by the simulation. This is then compared to unfiltered data, which includes noise, to test if this method of tracking can be implemented in the real experiment without access to the simulation information that performs this filtration. We find that
filtration by frequency is able to facilitate this. The real track patterns can then be isolated and included in the LUT.

The performance of this new tracking algorithm is thereafter evaluated by testing different such frequency thresholds. We discover that it is possible to create an algorithm with a LUT constructed by frequency filtration, and that it is able to match the performance of the existing algorithm for an optimal frequency threshold of 0.08% for a set of approximately one million events. This then establishes a tangible method for this algorithm to be implemented under real experimental conditions. (Less)
Please use this url to cite or link to this publication:
author
Kvarnstrom Meissner, Lucia LU
supervisor
organization
course
FYSK04 20261
year
type
M2 - Bachelor Degree
subject
language
English
id
9232752
date added to LUP
2026-06-09 10:10:23
date last changed
2026-06-09 10:10:23
@misc{9232752,
  abstract     = {{The Trigger Scintillator (TS) of the Light Dark Matter eXperiment (LDMX) is responsible for counting the electrons in the incoming beam in order to accurately determine the trigger threshold used in identifying missing momentum events. Electrons are counted by reconstructing their approximately horizontal tracks through the TS. The software algorithm operates on a maximum vertical tolerance value, making it both computationally expensive and inflexible to a misalignment in the TS geometry.

This thesis proposes an alternate counting algorithm using a data-driven lookup table (LUT) containing real electron patterns derived from simulation or data. This allows the tracking definition to naturally adapt to the TS geometry presented in the data. To write the LUT, electrons’ characteristic propagation patterns are first examined using truth-level information and then compared to unfiltered simulation data. It is shown that real tracks can be isolated from random combinatorics patterns by applying a minimum pattern frequency threshold to simulated data. At an optimal threshold of 0.0008, the data-driven LUT is able to achieve a maximum tracking efficiency 99.37% for a fake track rate 0.001668%, compared to the tolerance definition algorithm’s efficiency 98.00% and fake rate 0.00073%. This study demonstrates the feasibility of a data driven tracking method, with potential for further improvement through refinement of the frequency-threshold definition.}},
  author       = {{Kvarnstrom Meissner, Lucia}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Pattern-Based Electron Counting Algorithm for LDMX}},
  year         = {{2026}},
}