Test beam results for machine-learning-based reconstruction of resistive silicon sensors

Abstract

The advent of Low-Gain Avalanche Diodes (LGADs) and AC-coupled Low-Gain Avalanche Diodes (AC-LGADs) as candidate technologies for four-dimensional tracking has moved part of the reconstruction burden onto the analog waveforms of the sensor, in which the charge shared between electrodes encodes the impact position well below the electrode pitch. Recovering that position is a continuous, non-linear inverse problem. Transient Current Technique measurements characterize this problem at fixed injected charge and high signal-to-noise, whereas a minimum-ionizing beam adds Landau fluctuations in the deposited charge under detector operating conditions. We report test beam results for full-waveform reconstruction of an AC-LGAD of 500 μm pitch read out on a 2×2 pad matrix, exposed to the 120 GeV pion beam of the CERN Super Proton Synchrotron (SPS) and read out alongside a six-plane MIMOSA-26 telescope. The telescope is aligned, and its per-plane resolutions are calibrated in situ, giving a track pointing resolution of about 2.5 μm per axis at the device, so that the measured residual is the sensor response rather than a convolution with the reference. We then compare two analytic estimators, the parameter-free charge-imbalance formula and a fitted linear matrix inversion, with the topology-agnostic transformer developed for resistive silicon sensors. Over the full active pad, the transformer reaches a median total position with up to factor 3 improvement. Within the charge-sharing core, the three methods converge to the floor of this sensor at the achieved signal-to-noise.

Development and application of a machine learning potential for diamond surfaces and their sliding interfaces with silicon nitride and silica: effect of passivation, orientation and load
Enrico Pedretti, Maria Clelia Righi
arXiv28 Oct 2026PreprintCC BY 4.0Full textCondensed matterarXiv
Omnidirectional Radiation Detector with Perpendicular Dual Silicon Photomultiplier Readout - Directional Sensitivity and Machine Learning Source Positioning
Ana Marija Kožuljević, Gabriela Jazvac, Luka Lotina +1
arXiv6 Oct 2026PreprintCC BY 4.0Full textApplied physics & instrumentationarXiv
Status of Accelerator Automation in the Community: Classification, Survey Results, and the Role of AI and Machine Learning
S. Reimann, F. Miyahara, M. Lafky +10
arXiv7 Nov 2026PreprintCC BY 4.0Full textApplied physics & instrumentationarXiv
Enhancing electromagnetic calorimeter signal reconstruction with machine learning-based noise discrimination
Suman Das Gupta, Shamik Ghosh, Laltu Gazi +4
The European Physical Journal C30 Sep 2026Peer-reviewedCC BY 4.0Full textHigh-energy & particle physicsCrossref
Geometry-native machine learning reconstruction of DSMC moment fields with support monitoring
Ehsan Roohi
arXiv1 Sep 2026PreprintCC BY 4.0Full textComputational physics Fluid dynamicsarXiv

Cite

BibTeX RIS

BibTeX

@misc{wong2026test,
  title = {{Test beam results for machine-learning-based reconstruction of resistive silicon sensors}},
  author = {Don C. Wong and Leena Diehl and Spencer Ellis and Marcos Fernández García and Gabriele Giacomini and Vagelis Gkougkousis and Ben Kilminster and Jingyu Luo and Anna Macchiolo and Yajna Mahabeer and Yevhenii Padniuk and Daria Senina and Samantha Sunnarborg and Jessica Tang and Alessandro Tricoli and Lixing Wang and Gaetano Barone},
  howpublished = {arXiv},
  year = {2026},
  month = nov,
  doi = {10.48550/arxiv.2611.10194},
  eprint = {2611.10194},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2611.10194},
}

RIS

TY  - UNPB
TI  - Test beam results for machine-learning-based reconstruction of resistive silicon sensors
AU  - Don C. Wong
AU  - Leena Diehl
AU  - Spencer Ellis
AU  - Marcos Fernández García
AU  - Gabriele Giacomini
AU  - Vagelis Gkougkousis
AU  - Ben Kilminster
AU  - Jingyu Luo
AU  - Anna Macchiolo
AU  - Yajna Mahabeer
AU  - Yevhenii Padniuk
AU  - Daria Senina
AU  - Samantha Sunnarborg
AU  - Jessica Tang
AU  - Alessandro Tricoli
AU  - Lixing Wang
AU  - Gaetano Barone
PY  - 2026
DA  - 2026/11/06
PB  - arXiv
DO  - 10.48550/arxiv.2611.10194
UR  - https://arxiv.org/abs/2611.10194
AB  - The advent of Low-Gain Avalanche Diodes (LGADs) and AC-coupled Low-Gain Avalanche Diodes (AC-LGADs) as candidate technologies for four-dimensional tracking has moved part of the reconstruction burden onto the analog waveforms of the sensor, in which the charge shared between electrodes encodes the impact position well below the electrode pitch. Recovering that position is a continuous, non-linear inverse problem. Transient Current Technique measurements characterize this problem at fixed injected charge and high signal-to-noise, whereas a minimum-ionizing beam adds Landau fluctuations in the deposited charge under detector operating conditions. We report test beam results for full-waveform reconstruction of an AC-LGAD of 500 μm pitch read out on a 2×2 pad matrix, exposed to the 120 GeV pion beam of the CERN Super Proton Synchrotron (SPS) and read out alongside a six-plane MIMOSA-26 telescope. The telescope is aligned, and its per-plane resolutions are calibrated in situ, giving a track pointing resolution of about 2.5 μm per axis at the device, so that the measured residual is the sensor response rather than a convolution with the reference. We then compare two analytic estimators, the parameter-free charge-imbalance formula and a fitted linear matrix inversion, with the topology-agnostic transformer developed for resistive silicon sensors. Over the full active pad, the transformer reaches a median total position with up to factor 3 improvement. Within the charge-sharing core, the three methods converge to the floor of this sensor at the achieved signal-to-noise.
ER  -

Abstract

Related

Cite