Saved in:
Bibliographic Details
Main Authors: Brooke, James, Clement, Emyr, Glowacki, Maciej, Paramesvaran, Sudarshan, Segal, Jeronimo
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.09428
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913732908548096
author Brooke, James
Clement, Emyr
Glowacki, Maciej
Paramesvaran, Sudarshan
Segal, Jeronimo
author_facet Brooke, James
Clement, Emyr
Glowacki, Maciej
Paramesvaran, Sudarshan
Segal, Jeronimo
contents The implementation of convolutional neural networks in programmable logic, for applications in fast online event selection at hadron colliders is studied. In particular, an approach based on full event images for classification is studied, including hardware-aware optimisation of the network architecture, and evaluation of physics performance using simulated data. A range of network models are identified that can be implemented within resources of current FPGAs, as well as the stringent latency requirements of HL-LHC trigger systems. A candidate model that can be implemented in the CMS L1 trigger for HL-LHC was shown to be capable of excellent signal/background discrimination, although the performance depends strongly on the degree of pile-up mitigation possible prior to image generation.
format Preprint
id arxiv_https___arxiv_org_abs_2503_09428
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LHC Triggers using FPGA Image Recognition
Brooke, James
Clement, Emyr
Glowacki, Maciej
Paramesvaran, Sudarshan
Segal, Jeronimo
Instrumentation and Detectors
Computational Physics
The implementation of convolutional neural networks in programmable logic, for applications in fast online event selection at hadron colliders is studied. In particular, an approach based on full event images for classification is studied, including hardware-aware optimisation of the network architecture, and evaluation of physics performance using simulated data. A range of network models are identified that can be implemented within resources of current FPGAs, as well as the stringent latency requirements of HL-LHC trigger systems. A candidate model that can be implemented in the CMS L1 trigger for HL-LHC was shown to be capable of excellent signal/background discrimination, although the performance depends strongly on the degree of pile-up mitigation possible prior to image generation.
title LHC Triggers using FPGA Image Recognition
topic Instrumentation and Detectors
Computational Physics
url https://arxiv.org/abs/2503.09428