Speaker
Description
The ATLAS level-1 calorimeter trigger is a custom-built hardware system
that identifies events containing calorimeter-based physics objects,
including electrons, photons, taus, jets, and missing transverse energy.
In Run 3, L1Calo has been upgraded to process higher granularity
input data. The new trigger, currently running in parallel with the
legacy system, comprises several FPGA-based feature extractor modules,
which process the new digital information from the calorimeters and
execute more sophisticated trigger algorithms. The design of the
system will be presented along with an analysis of the improved
performance of the upgrade in the increasingly challenging Run-3
LHC pile-up environment.
Summary (500 words)
The ATLAS Level-1 Calorimeter provides the bulk of the hardware trigger
objects used to make the ATLAS level-1 trigger decision. Being based
on information from both the electromagnetic and hadronic layers of the
detector, it can identify potential electron, photon, tau and jet
candidates, as well as making a partial missing transverse energy assessment.
The original design, using over 7000 analogue inputs of coarse granularity
calorimeter information, was very successful in the first 10 years of
LHC operation, which already provided luminosities beyond design.
However, the expected higher luminosities in future LHC operation will
compromise the efficacy of the original hardware, and so an upgrade of
the system has been built and installed during the recent LHC shutdown
period. The basis of the improvement was to increase the level of detail
of information available to the trigger, with more granular information
both in longitudinal position, and calorimeter depth. In particular this
allows more sophisticated algorithms to be used based on shower shapes,
while also aiding energy resolution in a higher pile-up environment.
The higher data rate needed to transmit this additional information
(approximately a factor of 10 in the electromagnetic layer) necessitated
the use of a new digital trigger signal path which is integrated into
the calorimeter outputs, replacing the old analogue path with digital
signals transferred on optical links. The algorithmic part
of the Level-1 processing is performed on three Feature Extractors
(FEX), which specialize in identifying different physics signatures.
The electron feature extractor (eFEX) processes the full granularity
information in order to find smaller physics object showers,
using the full depth and spatial information to better distinguish, and
reject, the dominant jet background. It consists of 24 individual
ATCA-based modules running entirely independently and in parallel in
order to handle smaller blocks of the central detector coverage. The
jet feature extractor (jFEX) does not require the full granular
information, and so consists of only 6 similar modules, assessing
jet-like objects with a greater flexibility than the original
system for making regional corrections for pile-up effects. They
also provide a missing energy measurement. The global feature
extractor (gFEX) is a single module processing data from the whole
detector at a coarser granularity, also identifying jets and
measuring missing energy but, being a single module, has the capability
to make full event-level corrections.
The results from these FEX modules are further assessed by a new
topological trigger system, which can apply flexible algorithms from
simple multiplicity counting up to complex multiple object based
topological algorithms. These modules provide readout data via a
new custom ROD module and the ATLAS upgrade standard FELIX readout
mechanism. This data is required both to validate and monitor the
performance of the trigger, and as a seeding mechanism for the
ATLAS High Level Trigger.
The trigger is being switched over in stages from legacy to upgrade
during 2023, as validation of the correct functioning of the new
system is completed. Results on the rates and performance of the
upgrade trigger will be presented.
