AtlasPublicTopicHeader.png

L0MuonTriggerPublicResults

Introduction

Approved plots that can be shown by ATLAS speakers at conferences and similar events. Please do not add figures on your own. Contact the responsible project leader in case of questions and/or suggestions. Follow the guidelines on the trigger public results page. (Chrome & Firefox on MacOS have a problem to show the plots and it works for Safari.)

Level-0 MUON Trigger with multiple sub-systems

L0Muon trigger efficiency as a function of the truth muon pT (in the barrel region): ATL-COM-DAQ-2019-128 (14 Aug 2019)

Expected trigger efficiency for the Level-0 muon trigger with inclusions of MDT information (red) and without MDT (orange), compared with offline efficiency (green). The efficiency values relative to muons as trigger performance are obtained for muon tracks in the Large sectors in the barrel. The efficiency values are relative to a transverse momentum (pT) trigger threshold of 20 GeV. The values are obtained from single muon MC samples with no pile-up. Background level for the "Phase-II RPC" option is significantly higher than that for the "Phase-II RPC & MDT" option. .pdf
pdf


contact: Davide Cieri

Level-0 RPC Trigger

Convolutional Neural Network on FPGA for real time reconstruction of muons for the ATLAS Phase-II barrel trigger: ATL-COM-DAQ-2019-189 (17 Oct 2019)

An example input to a Convolutional Neural Network (CNN) for the Phase-2 ATLAS Level-0 muon trigger, implemented on a FPGA, is shown for a 13 GeV muon plus background. Resistive Plate Chambers (RPC) hits of a fixed $\phi$ sector are arranged in a matrix-like object. Each bin of the $y$-axis corresponds to a detector layer (3 detector layers for inner station, 4 for the middle and 2 for the outer station). The $x$-axis maps the $\eta$ coordinates of each physics RPC strip: for the $i$-th strip $\eta_{bin}^{i} = 384 \frac{\eta^{i}-\eta^{min}} {\eta^{max}-\eta^{min}}$, where $\eta^{max}$ and $\eta^{min}$ are respectively the maximum ($\eta^{max}=0.95$) and the minimum ($\eta^{min}=0.07$) $\eta$ values for the barrel RPC strips chosen to prevent muons to fall outside any layer of a specified sector; 384 is a realistic number of strips per layer. This particular choice has been taken in order to evaluate ML algorithm performances, without any geometrical acceptance effect. Random background has been added. The background rate has been evaluated from minimum bias events. Events used in the training phase of the CNN can also contain two or more muons in the same sector. Events with more than one muon are built superimposing one muon images with no background, which is then included. The CNN output is set to evaluate transverse momentum and $\eta$ of the leading and sub-leading muons (if the latter exists) in the sector and returns also a flag for events that contain more than 2 muons.

.pdf
pdf
contact: Luigi Sabetta & Federica Riti & Simone Francescato
An example input to a Convolutional Neural Network (CNN) for the Phase-2 ATLAS Level-0 muon trigger, implemented on a FPGA, is shown for background only. Resistive Plate Chambers (RPC) hits of a fixed $\phi$ sector are arranged in a matrix-like object. Each bin of the $y$-axis corresponds to a detector layer (3 detector layers for inner station, 4 for the middle and 2 for the outer station). The $x$-axis maps the $\eta$ coordinates of each physics RPC strip: for the $i$-th strip $\eta_{bin}^{i} = 384 \frac{\eta^{i}-\eta^{min}} {\eta^{max}-\eta^{min}}$, where $\eta^{max}$ and $\eta^{min}$ are respectively the maximum ($\eta^{max}=0.95$) and the minimum ($\eta^{min}=0.07$) $\eta$ values for the barrel RPC strips chosen to prevent muons to fall outside any layer of a specified sector; 384 is a realistic number of strips per layer. This particular choice has been taken in order to evaluate ML algorithm performances, without any geometrical acceptance effect. Random background has been added. The background rate has been evaluated from minimum bias events. Events used in the training phase of the CNN can also contain two or more muons in the same sector. Events with more than one muon are built superimposing one muon images with no background, which is then included. The CNN output is set to evaluate transverse momentum and $\eta$ of the leading and sub-leading muons (if the latter exists) in the sector and returns also a flag for events that contain more than 2 muons. .pdf
pdf
contact: Luigi Sabetta & Federica Riti & Simone Francescato
An example input to a Convolutional Neural Network (CNN) for the Phase-2 ATLAS Level-0 muon trigger, implemented on a FPGA, is shown for a 4 GeV muon without background. Resistive Plate Chambers (RPC) hits of a fixed $\phi$ sector are arranged in a matrix-like object. Each bin of the $y$-axis corresponds to a detector layer (3 detector layers for inner station, 4 for the middle and 2 for the outer station). The $x$-axis maps the $\eta$ coordinates of each physics RPC strip: for the $i$-th strip $\eta_{bin}^{i} = 384 \frac{\eta^{i}-\eta^{min}} {\eta^{max}-\eta^{min}}$, where $\eta^{max}$ and $\eta^{min}$ are respectively the maximum ($\eta^{max}=0.95$) and the minimum ($\eta^{min}=0.07$) $\eta$ values for the barrel RPC strips chosen to prevent muons to fall outside any layer of a specified sector; 384 is a realistic number of strips per layer. This particular choice has been taken in order to evaluate ML algorithm performances, without any geometrical acceptance effect. Random background has been added. The background rate has been evaluated from minimum bias events. Events used in the training phase of the CNN can also contain two or more muons in the same sector. Events with more than one muon are built superimposing one muon images with no background, which is then included. The CNN output is set to evaluate transverse momentum and $\eta$ of the leading and sub-leading muons (if the latter exists) in the sector and returns also a flag for events that contain more than 2 muons. .pdf
pdf
contact: Luigi Sabetta & Federica Riti & Simone Francescato
An example input to a Convolutional Neural Network (CNN) for the Phase-2 ATLAS Level-0 muon trigger, implemented on a FPGA, is shown for a 12 GeV and 15 GeV muons without background. Resistive Plate Chambers (RPC) hits of a fixed $\phi$ sector are arranged in a matrix-like object. Each bin of the $y$-axis corresponds to a detector layer (3 detector layers for inner station, 4 for the middle and 2 for the outer station). The $x$-axis maps the $\eta$ coordinates of each physics RPC strip: for the $i$-th strip $\eta_{bin}^{i} = 384 \frac{\eta^{i}-\eta^{min}} {\eta^{max}-\eta^{min}}$, where $\eta^{max}$ and $\eta^{min}$ are respectively the maximum ($\eta^{max}=0.95$) and the minimum ($\eta^{min}=0.07$) $\eta$ values for the barrel RPC strips chosen to prevent muons to fall outside any layer of a specified sector; 384 is a realistic number of strips per layer. This particular choice has been taken in order to evaluate ML algorithm performances, without any geometrical acceptance effect. Random background has been added. The background rate has been evaluated from minimum bias events. Events used in the training phase of the CNN can also contain two or more muons in the same sector. Events with more than one muon are built superimposing one muon images with no background, which is then included. The CNN output is set to evaluate transverse momentum and $\eta$ of the leading and sub-leading muons (if the latter exists) in the sector and returns also a flag for events that contain more than 2 muons. .pdf
pdf
contact: Luigi Sabetta & Federica Riti & Simone Francescato
An example input to a Convolutional Neural Network (CNN) for the Phase-2 ATLAS Level-0 muon trigger, implemented on a FPGA, is shown for a 13 GeV and 17 GeV muons with background. Resistive Plate Chambers (RPC) hits of a fixed $\phi$ sector are arranged in a matrix-like object. Each bin of the $y$-axis corresponds to a detector layer (3 detector layers for inner station, 4 for the middle and 2 for the outer station). The $x$-axis maps the $\eta$ coordinates of each physics RPC strip: for the $i$-th strip $\eta_{bin}^{i} = 384 \frac{\eta^{i}-\eta^{min}} {\eta^{max}-\eta^{min}}$, where $\eta^{max}$ and $\eta^{min}$ are respectively the maximum ($\eta^{max}=0.95$) and the minimum ($\eta^{min}=0.07$) $\eta$ values for the barrel RPC strips chosen to prevent muons to fall outside any layer of a specified sector; 384 is a realistic number of strips per layer. This particular choice has been taken in order to evaluate ML algorithm performances, without any geometrical acceptance effect. Random background has been added. The background rate has been evaluated from minimum bias events. Events used in the training phase of the CNN can also contain two or more muons in the same sector. Events with more than one muon are built superimposing one muon images with no background, which is then included. The CNN output is set to evaluate transverse momentum and $\eta$ of the leading and sub-leading muons (if the latter exists) in the sector and returns also a flag for events that contain more than 2 muons. .pdf
pdf
contact: Luigi Sabetta & Federica Riti & Simone Francescato
An example input to a Convolutional Neural Network (CNN) for the Phase-2 ATLAS Level-0 muon trigger, implemented on a FPGA, is shown. This example contains three muons without background. Resistive Plate Chambers (RPC) hits of a fixed $\phi$ sector are arranged in a matrix-like object. Each bin of the $y$-axis corresponds to a detector layer (3 detector layers for inner station, 4 for the middle and 2 for the outer station). The $x$-axis maps the $\eta$ coordinates of each physics RPC strip: for the $i$-th strip $\eta_{bin}^{i} = 384 \frac{\eta^{i}-\eta^{min}} {\eta^{max}-\eta^{min}}$, where $\eta^{max}$ and $\eta^{min}$ are respectively the maximum ($\eta^{max}=0.95$) and the minimum ($\eta^{min}=0.07$) $\eta$ values for the barrel RPC strips chosen to prevent muons to fall outside any layer of a specified sector; 384 is a realistic number of strips per layer. This particular choice has been taken in order to evaluate ML algorithm performances, without any geometrical acceptance effect. Random background has been added. The background rate has been evaluated from minimum bias events. Events used in the training phase of the CNN can also contain two or more muons in the same sector. Events with more than one muon are built superimposing one muon images with no background, which is then included. The CNN output is set to evaluate transverse momentum and $\eta$ of the leading and sub-leading muons (if the latter exists) in the sector and returns also a flag for events that contain more than 2 muons. .pdf
pdf
contact: Luigi Sabetta & Federica Riti & Simone Francescato
An example input to a Convolutional Neural Network (CNN) for the Phase-2 ATLAS Level-0 muon trigger, implemented on a FPGA, is shown. This example contains three muons with background. Resistive Plate Chambers (RPC) hits of a fixed $\phi$ sector are arranged in a matrix-like object. Each bin of the $y$-axis corresponds to a detector layer (3 detector layers for inner station, 4 for the middle and 2 for the outer station). The $x$-axis maps the $\eta$ coordinates of each physics RPC strip: for the $i$-th strip $\eta_{bin}^{i} = 384 \frac{\eta^{i}-\eta^{min}} {\eta^{max}-\eta^{min}}$, where $\eta^{max}$ and $\eta^{min}$ are respectively the maximum ($\eta^{max}=0.95$) and the minimum ($\eta^{min}=0.07$) $\eta$ values for the barrel RPC strips chosen to prevent muons to fall outside any layer of a specified sector; 384 is a realistic number of strips per layer. This particular choice has been taken in order to evaluate ML algorithm performances, without any geometrical acceptance effect. Random background has been added. The background rate has been evaluated from minimum bias events. Events used in the training phase of the CNN can also contain two or more muons in the same sector. Events with more than one muon are built superimposing one muon images with no background, which is then included. The CNN output is set to evaluate transverse momentum and $\eta$ of the leading and sub-leading muons (if the latter exists) in the sector and returns also a flag for events that contain more than 2 muons. .pdf
pdf
contact: Luigi Sabetta & Federica Riti & Simone Francescato
A Convolutional Neural Network (CNN) implementation for the Phase-2 ATLAS Level-0 muon trigger with floating point weights has been set up and trained. The inputs to this network are all the strips of all the detector layers of a sector. The predicted leading muon transverse momentum ($p_T$) is shown as a function of the leading muon $p_T$ after detector simulation (true $p_T$ ). The reconstructed transverse momentum is used for the network training phase. The columns of the histogram are normalized to unity, since leading muon $p_T$ distribution is not flat. .pdf
pdf
contact: Luigi Sabetta & Federica Riti & Simone Francescato
A ternary Convolutional Neural Network (tCNN, arXiv:1605.04711) implementation for the Phase-2 ATLAS Level-0 muon trigger has been set up and trained. The inputs to this network are all the strips of all the detector layers of a sector. The predicted leading muon transverse momentum ($p_T$ ) is shown as a function of the leading muon $p_T$ after detector simulation (true $p_T$ ). Ternary networks have weights made of just 2 bits. For this reason, tCNNs represent an optimal solution for FPGA synthesis and implementation, since the resource occupancy can be reduced up to a factor of 16, compared to a same-architecture non-ternary network. The columns of the histogram are normalized to unity. .pdf
pdf
contact: Luigi Sabetta & Federica Riti & Simone Francescato
A ternary Convolutional Neural Network (tCNN, arXiv:1605.04711) implementation for the Phase-2 ATLAS Level-0 muon trigger has been set up and trained. The inputs to this network are all the strips of all the detector layers of a sector. The predicted number of muon is shown as a function of the numbers of muons obtained from truth information after detector simulation. Ternary networks have weights made of just 2 bits. For this reason, tCNNs represent an optimal solution for FPGA synthesis and implementation, since the resource occupancy can be reduced up to a factor of 16, compared to a same-architecture non-ternary network. The columns of the histogram are normalized to unity. No $p_T$ selection is applied to muon candidates. Fake muons rate is at the level of 0.6% but is furthermore reduced requiring reasonable $p_T$ thresholds (great fraction of fakes is predicted at very low momentum). .pdf
pdf
contact: Luigi Sabetta & Federica Riti & Simone Francescato
A comparison of the efficiencies of the Phase-2 ATLAS Level-0 standard muon trigger algorithm (cyan), the Convolutional Neural Network (red, CNN) and the ternary Convolutional Neural Network (blue, tCNN) is shown as a function of the transverse momentum $p_T$ . For the standard algorithm only the efficiency obtained with the configuration which requires at least 3 spatial coincidences over the 4 Resistive Plate Chambers stations is shown. All the efficiency curves take the single-muon images without background as input, and are tuned so that the efficiency reached at nominal threshold is the same (82%) to ease the comparison. .pdf
pdf
contact: Luigi Sabetta & Federica Riti & Simone Francescato
The efficiencies of the Phase-2 ATLAS Level-0 standard muon trigger algorithm (cyan) and the Convolutional Neural Networks with ternary weights (tCNN) and varying input sizes is shown as a function of the transverse momentum $p_T$ . For the standard algorithm only the efficiency obtained with the configuration which requires at least 3 spatial coincidences over the 4 Resistive Plate Chambers stations is shown. The network which takes as input all the strips of the images and has the best performance (blue) is shown. In magenta, a tCNN which takes as input just a section of the whole image at a time, and operate scanning the image, is shown. Passing a smaller input to the tCNN widely reduces the total number of multiplication, therefore making it possible to increase the parallelization and to fullfill the latency requirement. All the efficiency curves take the single-muon images without background as input, and are tuned so that the efficiency reached at nominal threshold is the same (82%) to ease the comparison. .pdf
pdf
contact: Luigi Sabetta & Federica Riti & Simone Francescato

Level-0 TGC Trigger

Performance of trigger algorithms in firmware simulation: ATL-COM-DAQ-2022-070 (25 Aug 2022)

Distribution of the polar angle difference Δθ between a track segment reconstructed by a pattern matching algorithm with Thin Gap Chamber (TGC) hits and a truth track segment. Precise measurement of the track segment angle is crucial in determining muon transverse momentum for trigger. The pattern matching algorithm for a full coverage of TGC is implemented on Virtex UltraScale+ FPGA XCVU13P in Vivado firmware simulation. The TGC hits are obtained from ATLAS Geant4 Monte-Carlo (MC) sample and used as the simulation inputs. The output of the simulation is the track segment angle. The width of the Δθ distribution is regarded as resolution of the TGC pattern matching algorithm. The MC sample includes a muon in each event, and the events are selected by requiring the muon emitted from the interaction point in a pseudorapidity range 1.05 < |η| < 2.4. .pdf
pdf png eps
contact: Yuki Mitsumori & Yasuyuki Horii
Expected angular resolution of track segment reconstruction with Thin Gap Chamber (TGC) pattern matching algorithm as a function of the magnitude of the truth muon pseudorapidity (η). The angular resolution is obtained in firmware simulation with TGC pattern matching algorithm implemented on Virtex UltraScale+ FPGA XCVU13P. The input of the simulation is provided from Monte-Carlo sample including a muon in each event, where the transverse momentum is random in 20-100 GeV. The angular resolution depends on |η| due to difference in channel sizes of TGC. .pdf
pdf png eps
contact: Yuki Mitsumori & Yasuyuki Horii
Expected efficiency of track segment reconstruction with Thin Gap Chamber (TGC) pattern matching algorithm as a function of the magnitude of the truth muon pseudorapidity (η). The efficiency is obtained in firmware simulation with TGC pattern matching algorithm implemented on Virtex UltraScale+ FPGA XCVU13P. The input of the simulation is provided from Monte-Carlo sample including a muon in each event, where the transverse momentum is random in 20-100 GeV. The efficiency is relatively low in 1.05 < |η| < 1.1 due to limited acceptance of TGC detector in the transition region between the barrel and the endcap. Relatively low efficiencies at |η| ~ 1.4 and 2.2 are due to the holes of the TGC detectors, which are provided for the laser paths of the alignment system of the Monitored Drift Tube detectors. .pdf
pdf png eps
contact: Yuki Mitsumori & Yasuyuki Horii
Expected efficiency of track segment reconstruction with Thin Gap Chamber (TGC) pattern matching algorithm as a function of the truth muon azimuthal angle (φ). The efficiency is obtained in firmware simulation with TGC pattern matching algorithm implemented on Virtex UltraScale+ FPGA XCVU13P. The input of the simulation is provided from Monte-Carlo sample including a muon in each event. The muon has transverse momentum 20-100 GeV in a pseudorapidity range 1.05 < |η| < 2.4. The efficiency is relatively low in one out of three bins in the whole φ region due to the holes of the TGC detectors, which are provided for the laser paths of the alignment system of the Monitored Drift Tube detectors. .pdf
pdf png eps
contact: Yuki Mitsumori & Yasuyuki Horii

Performance of trigger algorithms in software and firmware implementations: ATL-COM-DAQ-2019-081 (25 Jul 2019)

Expected efficiency for the Level-0 muon trigger with HL-LHC scheme (red) and with LHC Run-2 scheme (blue) . The efficiency values relative to offline muons as trigger performance are obtained for a pseudorapidity range 1.05 < |η| < 2.4 and a transverse momentum (pT) threshold of 20 GeV with a single muon Monte Carlo simulation sample for the HL-LHC scheme, and data taken in 2018 for the Run-2 scheme. The HL-LHC scheme is based on Thin Gap Chamber (TGC), Tile Calorimeter (TileCal) and New Small Wheel (NSW). The HL-LHC scheme assumes track segment reconstruction with TGC hits by a pattern matching algorithm combined with the TileCal and NSW information for the determination of pT. The segments from Monitored Drift Tube are used to emulate the NSW segments. The HL-LHC scheme provides a higher efficiency in the plateau region with better rejection of low pT muons. The high efficiency at the plateau in the HL-LHC scheme comes from a loose condition for the coincidence among the TGC layers in the pattern matching algorithm, which requires 5 hits in 7 layers. The better pT resolution in the turn-on curve is from improvements in the TGC track reconstruction algorithm. .pdf
pdf
contact: Yuya Mino & Toshi Sumida
Estimated rate of the Level-0 single muon trigger at HL-LHC based on Thin Gap Chamber, Tile Calorimeter, and New Small Wheel for a pseudorapidity range 1.05 < |η| < 2.4 and a transverse momentum threshold of 20 GeV. The segments from Monitored Drift Tube are used to emulate the NSW segments. Events in the Run-2 data sample taken in 2016 with the zero-bias trigger are used for the point with 1x1034 cm-2s-1 and overlaid to account for higher luminosity points. The number of the simultaneous interactions in each bunch crossing ("pileup") in the original data is 27 in average. Pileup conditions at higher luminosity values are simulated for each event by overlaying a number of zero-bias events drawn from the expected distribution for that luminosity. The solid line shows a fitting result by a linear function crossing the origin. The dashed lines show the trigger rate at the luminosity in the HL-LHC (7.5x1034 cm-2s-1), calculated from a nominal number of pileup (200). .png
pdf
contact: Yuya Mino & Toshi Sumida
Distribution of the polar angle difference Δθ between a track segment reconstructed by a pattern matching algorithm with Thin Gap Chamber (TGC) hits assumed to be used in the Level-0 muon trigger at HL-LHC and a track segment reconstructed by an offline algorithm based on Monitored Drift Tube (MDT) hits. Precise measurement of the track segment angle is profited to determine muon transverse momentum and obtain sharper efficiency turn-on curves. The TGC pattern matching algorithm is implemented in a Virtex UltraScale+ FPGA XCVU9P on an evaluation kit. Test vectors of TGC hits are obtained from Monte-Carlo (MC) sample and used as the FPGA inputs. The FPGA outputs are compared with the offline track segments based on MDT hits. The offline track segment is considered to be a reference, thus the width of Δθ distribution is regarded as resolution of the TGC pattern matching algorithm for muon tracks used in this study. The MC sample includes a muon in each event, and the events are selected by requiring exactly seven TGC hits on the seven layers, i.e. neither missing hits nor cross talks, in a pseudorapidity range 2.13 < η < 2.16. Memory usage of the algorithm in this study corresponds to about one third of the total memory resource of XCVU9P when a full η range of the endcap is included. .png
pdf
contact: Haruka Asada & Yasuyuki Horii

Level-0 MDT Trigger

Phase-II L0MDT performance plots: ATL-COM-DAQ-2020-110 (29 Mar 2021)

The muon pT resolution σ as a function of the offline muon pT, where σ is defined as the standard deviation from a Gaussian fit to the pT imbalance between the muon pT reconstructed from track segments in the muon spectrometer and the offline muon pT, divided by the offline muon pT. The values are obtained from single muon MC samples with no pile-up or cavern background. When three MDT track segments are reconstructed the "sagitta" method is used, otherwise, the "deflection angle" method is used. .pdf
pdf
contact: Mats Joakim Robert Olsson
The muon pT resolution σ as a function of the offline muon pseudorapidity (η), where σ is defined as the standard deviation from a Gaussian fit to the pT imbalance between the muon pT reconstructed from track segments in the muon spectrometer and the offline muon pT, divided by the offline muon pT. The values are obtained from single muon MC samples with no pile-up or cavern background. When three MDT track segments are reconstructed the "sagitta" method is used, otherwise, the "deflection angle" method is used. The significant deterioration in σ for 1.1 < η < 1.8 is due to the more complicated magnetic field in the "transition region" between the barrel and the endcap. The deflection angle method performs significantly worse than the sagitta method in the forward region, due to the weaker magnetic field here. .pdf
pdf
contact: Mats Joakim Robert Olsson
The muon pT resolution σ as a function of the offline muon azimuthal angle (Φ), where σ is defined as the standard deviation from a Gaussian fit to the pT imbalance between the muon pT reconstructed from track segments in the muon spectrometer and the offline muon pT, divided by the offline muon pT. The values are obtained from single muon MC samples with no pile-up or cavern background. When three MDT track segments are reconstructed the "sagitta" method is used, otherwise, the "deflection angle" method is used. Small deteriorations in σ are observed for sectors in the "feet" region (Φ<0). .pdf
pdf
contact: Mats Joakim Robert Olsson
The muon pT resolution σ as a function of the offline muon pT, where is defined as the standard deviation from a Gaussian fit to the pT imbalance between the muon pT reconstructed from track segments in the muon spectrometer and the offline muon pT, divided by the offline muon pT. The values are obtained from single muon MC samples with no pile-up or cavern background. In order to account for the non-uniform magnetic field, corrections with respect to the relative azimuthal angle (Φ_mod = Φ - Φ_region center) and pseudorapidity (η) are performed. The muon pT resolution is shown with no correction, when adding the Φ_mod correction, and when adding both the Φ_mod and the η corrections. The Φ_mod correction significantly improves the pT resolution. .pdf
pdf
contact: Mats Joakim Robert Olsson
The muon pT resolution σ as a function of the offline muon pseudorapidity (η), where σ is defined as the standard deviation from a Gaussian fit to the pT imbalance between the muon pT reconstructed from track segments in the muon spectrometer and the offline muon pT, divided by the offline muon pT. The values are obtained from single muon MC samples with no pile-up or cavern background. In order to account for the non-uniform magnetic field, corrections with respect to the relative azimuthal angle (Φ_mod = Φ - Φ_region center) and pseudorapidity (η) are performed. The muon pT resolution is shown with no correction, when adding the Φ_mod correction, and when adding both the Φ_mod and the η corrections. The Φ_mod correction significantly improves the pT resolution. The significant deterioration in σ for 1.1 < η < 1.8 is due to the more complicated magnetic field in the "transition region" between the barrel and the endcap. .pdf
pdf
contact: Mats Joakim Robert Olsson
The muon pT resolution σ as a function of the offline azimuthal angle (Φ), where σ is defined as the standard deviation from a Gaussian fit to the pT imbalance between the muon pT reconstructed from track segments in the muon spectrometer and the offline muon pT, divided by the offline muon pT. The values are obtained from single muon MC samples with no pile-up or cavern background. In order to account for the non-uniform magnetic field, corrections with respect to the relative azimuthal angle (Φ_mod = Φ - Φ_region center) and pseudorapidity (η) are performed. The muon pT resolution is shown with no correction, when adding the Φ_mod correction, and when adding both the Φ_mod and the η corrections. Small deteriorations in σ are observed for sectors in the "feet" region (Φ < 0). .pdf
pdf
contact: Mats Joakim Robert Olsson

An extension of Associative Memory approach to tracking with a drift-tube detector using timing information and its demonstration for HL-LHC ATLAS muon trigger: ATL-COM-DAQ-2020-106 (20 Oct 2020)

The definitions of three observables (α, β, sagitta (s)) that are used in the demonstration study of online muon reconstruction with the Associative Memory approach. Muon spectrometer consists of three stations (Inner, Middle, and Outer) in the Barrel region, and the observables can be evaluated using the straight track-segments in the three stations. They are proxies for the curvature of muon trajectories in the troidal magnetic field and approximately proportional to the inverse of transverse momentum (pT) of muons. The observable of α is defined as the angle difference between the direction of the segments and straight trajectories pointing the interaction points, corresponding to trajectories with infinite momentum; the β is defined as the angular difference of the direction of segments in two different MDT stations; the sagitta is defined as the distance from the position of the segment in the middle MDT station to a virtual straight muon track drawn between the segments in the inner and the outer MDT stations. The three observables can be used in the online pT selection in triggering high pT muons. The sagitta provides the best resolution in the pT estimation of the three while the β and α mitigate the effect of inefficiency in the straight track-segment reconstruction. .pdf
pdf
contact: Yasu Okumura; Yunjian He
The relative efficiency of the refined transverse momentum (pT) selection provided by the online reconstruction using Associative Memory (AM) approach with reference to the offline muons that passes the medium quality requirement and are matched to the Run-2 Level-1 RPC trigger RoI in the barrel region satisfying pT threshold of 20 GeV (L1 MU20). The refined selection using the online muon reconstruction are applied in order to improve the momentum resolution (Refined L1 MU20). The Run-2 L1 MU20 candidates are collected by pass-through triggers (HLT noalg L1MU20) in 2018 data with a centre-of-mass energy of 13 TeV and a bunch-crossing interval of 25 nsec. The online reconstruction with the AM approach provides additional rejection capability for muons with pT < 20 GeV while it keeps high efficiency for high pT muons with a plateau efficiency 99.5%. The efficiency rises for muons below 6 GeV as the system-design and AM pattern list is optimised for muon with pT > 6 GeV in this demonstration study. The offline pT measured by the muon spectrometer with a constant offset correction corresponding to the energy deposit in the materials in front of the muon spectrometer (2.8 GeV) is taken as the x-axis in the figure. .pdf
pdf
contact: Yasu Okumura; Yunjian He
The transverse momentum (pT) distribution of offline reconstructed muons matched to a Level-1 RoI of a single muon with pT threshold at 20 GeV (L1 MU20) with |ηRoI| < 1.05, which are used in the demonstration study of the online muon reconstruction with the Associative Memory (AM) approach. The medium quality is required in the offline selection. Matching between the offline muon and the L1 MU20 RoI requires |∆η| < 0.08 and |∆φ| < 0.18, where |∆η| and |∆φ| are calculated from η and φ of the offline-reconstructed muon and the central position of the L1 MU20 RoI. The Run-2 L1 MU20 candidates are collected by pass-through triggers (HLT noalg L1MU20) in 2018 data with a centre-of-mass energy of 13 TeV and a bunch-crossing interval of 25 nsec. The straight track-segment reconstruction with the AM approach is simulated at each station on the muon trajectories for the demonstration of the online reconstruction. The muons are classified into five categories with respect to the number of available online track-segments for the momentum determination. About 90% of the muons have online-segments reconstructed at the all three stations (“3 stations”) and can use the sagitta method in the online momentum determination. Muons categorised into “2 stations” and “Middle/Outer Only” (about 10%) can be reconstructed by the β method and the α method, respectively, and these methods mitigate potential inefficiency in the track-segment reconstruction stage. The remaining small fraction of muons (less than 0.15%) fail the reconstruction of enough number of online track-segments to estimate the momentum, categorized into either of “Inner Only” or “0 station” in the figure. The offline pT measured by the muon spectrometer with a constant offset correction corresponding to the energy deposit in the materials in front of the muon spectrometer (2.8 GeV) is taken as the x-axis in the figure. .pdf
pdf
contact: Yasu Okumura; Yunjian He
The transverse momentum (pT) distribution of offline reconstructed muons matched to a Level-1 RoI of a single muon with pT threshold at 20 GeV (L1 MU20) with |ηRoI| < 1.05, which are used in the demonstration study of the online muon reconstruction with the Associative Memory (AM) approach. The medium quality is required in the offline selection. Matching between the offline muon and the L1 MU20 RoI requires |∆η| < 0.08 and |∆φ| < 0.18, where |∆η| and |∆φ| are calculated from η and φ of the offline-reconstructed muon and the central position of the L1 MU20 RoI. The Run-2 L1 MU20 candidates are collected by pass-through triggers (HLT noalg L1MU20) in 2018 data with a centre-of-mass energy of 13 TeV and a bunch-crossing interval of 25 nsec. The online muon reconstruction with the AM approach is simulated to provide a refined selectivity in the triggering. The filled histogram with blue colour shows the pT distribution of muons that satisfy the pT threshold for 20 GeV trigger with the refined momentum information provided by the online muon reconstruction with AM approach, denoted as “Refined L1 MU20” in the figure. It provides additional rejection capability for muons with pT < 20 GeV while it keeps high efficiency for high pT muons with a plateau efficiency 99.5%. The offline pT measured by the muon spectrometer with a constant offset correction corresponding to the energy deposit in the materials in front of the muon spectrometer (2.8 GeV) is taken as the x-axis in the figure. .pdf
pdf
contact: Yasu Okumura; Yunjian He
The relative trigger rate with respect to the Run-2 L1 MU20 as a function of the average number of interactions per bunch crossing (⟨μ⟩, pileup). The refined selection using muon transverse momentum (pT) provided by the online muon reconstruction with the Associative Memory (AM) (Refined L1 MU20) approach are applied to Run-2 Level-1 RoIs of a single muon with pT threshold at 20 GeV (Run-2 L1 MU20) with |ηRoI| < 1.05. The Run-2 L1 MU20 candidates are collected by pass-through triggers (HLT noalg L1MU20) in 2018 data with a centre-of-mass energy of 13 TeV and a bunch-crossing interval of 25 nsec. Additional 50% rate reduction can be provided by the refined selection, and the performance does not depend on the pileup in the range up to ⟨μ⟩ of 50. In addition, the figure shows 60% of muons with pT < 20 GeV are rejected by the refined selection given by the online muon reconstruction with the AM approach. .pdf
pdf
contact: Yasu Okumura; Yunjian He
Online straight track-segment reconstruction with the Associative Memory approach runs for offline-reconstructed straight track-segments at the three muon stations in the Barrel region, and the residual of angle θ of straight track-segments between the offline and the online reconstruction. The pattern-finding algorithm with the AM approach will find offline segments for given sets of MDT hits at three muon stations, and representative θ values of AM patterns (θonline) are used as the online estimations. For the study, Z → μμ processes are simulated with the background environment expected for the High-Luminosity LHC (HL-LHC). The figure shows the residual of θ at the barrel inner station in with the corresponding angle resolution of σ = 0.95 mrad. .pdf
pdf
contact: Yasu Okumura; Yunjian He
Online straight track-segment reconstruction with the Associative Memory approach runs for offline-reconstructed straight track-segments at the three muon stations in the Barrel region, and the residual of position z of straight track-segments between the offline and the online reconstruction. The pattern-finding algorithm with the AM approach will find offline segments for given sets of Monitored Drift Tube hits at three muon stations, and representative z values of AM patterns (zonline) are used as the online estimations. For the study, Z → μμ processes are simulated with the background environment expected for the High-Luminosity LHC (HL-LHC). The figure shows the residual of z at the barrel inner station with the corresponding position resolution of σ = 200 μm.

.pdf
pdf
contact: Yasu Okumura; Yunjian He
The number of the segment patterns to be stored in Associative Memory per operation unit as a function of “Don’t-Care” (DC) bit configurations (Nmax) [1]. The DC DC bit application allows us to have a variable resolution that is configurable for individual patterns and layers in the drift-radius representation. The capability reduces the number of patterns, while minimising the impact on the resolutions and the fake matching rate. The Nmax is the maximum allowed number of DC max DC bits per straight track-segment pattern, and various configurations of the NDC is tested to search for optimal configurations. The DC bit configuration that satisfies the pattern-bank size requirement while minimising Nmax is chosen as the optimal solution for each station in the pattern training. The optimisation DC study is repeated for the five MDT stations (Barrel Inner, Barrel Middle, Barrel Outer, Endcap Middle, and Endcap Outer). As the number of MDT tubes are different for individual stations, the needed pattern-bank sizes are different with a given DC bit configuration for different sectors. The two horizontal dashed lines show the limit of the pattern-bank size as a reference. They are 800k for Barrel Inner, Barrel Middle, and Endcap Outer and 1,600k for Barrel Outer and Endcap Middle as an optimal configuration. In the studies, four bits are assigned to the drift-radius representation in the Content-addressable Memory (CAM) in the Associative Memory device, corresponding to 1 mm binning. Among four bits, up to two bits can be treated as DC bits in a configurable manner. As one pattern consists of six hits, the Nmax is varied from zero to 12 as shown in the figure. The application of the DC bit allows us to find solutions that satisfy the DC limitation for all the detector regions. Reference: [1] ATLAS Collaboration, Fast TracKer (FTK) Technical Design Report, CERN-LHCC-2013-007, ATLAS-TDR-021, https://cds.cern.ch/record/1552953 .pdf
pdf
contact: Yasu Okumura; Yunjian He

Major updates:
-- MasayaIshino - 2019-07-24

Responsible: MasayaIshino
Subject: Phase2 Upgrade Trigger

Attachments Attachments
Topic attachments
I Attachment History Action Size Date Who Comment
PDFpdf ATL-COM-DAQ-2019-081-fig1.pdf r1 manage 36.4 K 2019-09-20 - 17:36 MasayaIshino  
PNGpng ATL-COM-DAQ-2019-081-fig1.png r1 manage 130.5 K 2020-10-30 - 06:02 MasayaIshino  
PDFpdf ATL-COM-DAQ-2019-081-fig2.pdf r1 manage 28.9 K 2019-09-20 - 17:36 MasayaIshino  
PNGpng ATL-COM-DAQ-2019-081-fig2.png r1 manage 117.8 K 2020-10-30 - 06:02 MasayaIshino  
PDFpdf ATL-COM-DAQ-2019-081-fig3.pdf r1 manage 24.9 K 2019-09-20 - 17:36 MasayaIshino  
PNGpng ATL-COM-DAQ-2019-081-fig3.png r1 manage 90.4 K 2020-10-30 - 06:02 MasayaIshino  
PDFpdf ATL-COM-DAQ-2019-128-fig1.pdf r1 manage 16.8 K 2019-09-20 - 17:49 MasayaIshino  
PNGpng ATL-COM-DAQ-2019-128-fig1.png r1 manage 109.4 K 2020-10-30 - 05:59 MasayaIshino  
PDFpdf ATL-COM-DAQ-2019-189-fig1.pdf r1 manage 173.2 K 2019-11-08 - 10:57 RiccardoVari  
PNGpng ATL-COM-DAQ-2019-189-fig1.png r1 manage 46.5 K 2020-10-30 - 06:07 MasayaIshino  
PDFpdf ATL-COM-DAQ-2019-189-fig10.pdf r1 manage 71.7 K 2019-11-08 - 11:16 RiccardoVari  
PNGpng ATL-COM-DAQ-2019-189-fig10.png r1 manage 75.1 K 2020-10-30 - 06:07 MasayaIshino  
PDFpdf ATL-COM-DAQ-2019-189-fig11.pdf r1 manage 108.3 K 2019-11-08 - 11:16 RiccardoVari  
PNGpng ATL-COM-DAQ-2019-189-fig11.png r1 manage 74.3 K 2020-10-30 - 06:08 MasayaIshino  
PDFpdf ATL-COM-DAQ-2019-189-fig12.pdf r1 manage 108.7 K 2019-11-08 - 11:16 RiccardoVari  
PNGpng ATL-COM-DAQ-2019-189-fig12.png r1 manage 76.3 K 2020-10-30 - 06:08 MasayaIshino  
PDFpdf ATL-COM-DAQ-2019-189-fig2.pdf r1 manage 172.5 K 2019-11-08 - 11:01 RiccardoVari  
PNGpng ATL-COM-DAQ-2019-189-fig2.png r1 manage 45.2 K 2020-10-30 - 06:07 MasayaIshino  
PDFpdf ATL-COM-DAQ-2019-189-fig3.pdf r1 manage 172.3 K 2019-11-08 - 11:01 RiccardoVari  
PNGpng ATL-COM-DAQ-2019-189-fig3.png r1 manage 44.6 K 2020-10-30 - 06:07 MasayaIshino  
PDFpdf ATL-COM-DAQ-2019-189-fig4.pdf r1 manage 172.5 K 2019-11-08 - 11:01 RiccardoVari  
PNGpng ATL-COM-DAQ-2019-189-fig4.png r1 manage 45.6 K 2020-10-30 - 06:07 MasayaIshino  
PDFpdf ATL-COM-DAQ-2019-189-fig5.pdf r1 manage 172.9 K 2019-11-08 - 11:16 RiccardoVari  
PNGpng ATL-COM-DAQ-2019-189-fig5.png r1 manage 46.7 K 2020-10-30 - 06:07 MasayaIshino  
PDFpdf ATL-COM-DAQ-2019-189-fig6.pdf r1 manage 172.7 K 2019-11-08 - 11:16 RiccardoVari  
PNGpng ATL-COM-DAQ-2019-189-fig6.png r1 manage 46.6 K 2020-10-30 - 06:07 MasayaIshino  
PDFpdf ATL-COM-DAQ-2019-189-fig7.pdf r1 manage 173.2 K 2019-11-08 - 11:16 RiccardoVari  
PNGpng ATL-COM-DAQ-2019-189-fig7.png r1 manage 46.9 K 2020-10-30 - 06:07 MasayaIshino  
PDFpdf ATL-COM-DAQ-2019-189-fig8.pdf r1 manage 78.6 K 2019-11-08 - 11:16 RiccardoVari  
PNGpng ATL-COM-DAQ-2019-189-fig8.png r1 manage 86.6 K 2020-10-30 - 06:07 MasayaIshino  
PDFpdf ATL-COM-DAQ-2019-189-fig9.pdf r1 manage 78.4 K 2019-11-08 - 11:16 RiccardoVari  
PNGpng ATL-COM-DAQ-2019-189-fig9.png r1 manage 76.8 K 2020-10-30 - 06:07 MasayaIshino  
PDFpdf ATL-COM-DAQ-2020-106-fig1.pdf r2 r1 manage 43.8 K 2020-10-30 - 13:39 MasayaIshino  
PNGpng ATL-COM-DAQ-2020-106-fig1.png r2 r1 manage 179.4 K 2020-10-30 - 13:45 MasayaIshino  
PDFpdf ATL-COM-DAQ-2020-106-fig2.pdf r2 r1 manage 27.0 K 2020-10-30 - 13:40 MasayaIshino  
PNGpng ATL-COM-DAQ-2020-106-fig2.png r2 r1 manage 161.3 K 2020-10-30 - 13:46 MasayaIshino  
PDFpdf ATL-COM-DAQ-2020-106-fig3.pdf r2 r1 manage 31.7 K 2020-10-30 - 13:40 MasayaIshino  
PNGpng ATL-COM-DAQ-2020-106-fig3.png r2 r1 manage 148.6 K 2020-10-30 - 13:48 MasayaIshino  
PDFpdf ATL-COM-DAQ-2020-106-fig4.pdf r2 r1 manage 25.4 K 2020-10-30 - 13:40 MasayaIshino  
PNGpng ATL-COM-DAQ-2020-106-fig4.png r2 r1 manage 144.6 K 2020-10-30 - 13:49 MasayaIshino  
PDFpdf ATL-COM-DAQ-2020-106-fig5.pdf r2 r1 manage 35.8 K 2020-10-30 - 13:41 MasayaIshino  
PNGpng ATL-COM-DAQ-2020-106-fig5.png r2 r1 manage 184.7 K 2020-10-30 - 13:51 MasayaIshino  
PDFpdf ATL-COM-DAQ-2020-106-fig6.pdf r2 r1 manage 32.4 K 2020-10-30 - 13:41 MasayaIshino  
PNGpng ATL-COM-DAQ-2020-106-fig6.png r2 r1 manage 154.1 K 2020-10-30 - 13:52 MasayaIshino  
PDFpdf ATL-COM-DAQ-2020-106-fig7.pdf r2 r1 manage 31.9 K 2020-10-30 - 13:41 MasayaIshino  
PNGpng ATL-COM-DAQ-2020-106-fig7.png r2 r1 manage 153.6 K 2020-10-30 - 13:53 MasayaIshino  
PDFpdf ATL-COM-DAQ-2020-106-fig8.pdf r2 r1 manage 25.2 K 2020-10-30 - 13:41 MasayaIshino  
PNGpng ATL-COM-DAQ-2020-106-fig8.png r2 r1 manage 211.9 K 2020-10-30 - 13:54 MasayaIshino  
PDFpdf ATL-COM-DAQ-2020-110-fig1.pdf r1 manage 28.7 K 2021-03-30 - 04:03 MasayaIshino  
PNGpng ATL-COM-DAQ-2020-110-fig1.png r1 manage 107.5 K 2021-03-30 - 04:06 MasayaIshino  
PDFpdf ATL-COM-DAQ-2020-110-fig2.pdf r1 manage 28.7 K 2021-03-30 - 04:03 MasayaIshino  
PNGpng ATL-COM-DAQ-2020-110-fig2.png r1 manage 108.2 K 2021-03-30 - 04:06 MasayaIshino  
PDFpdf ATL-COM-DAQ-2020-110-fig3.pdf r1 manage 26.6 K 2021-03-30 - 04:03 MasayaIshino  
PNGpng ATL-COM-DAQ-2020-110-fig3.png r1 manage 98.9 K 2021-03-30 - 04:06 MasayaIshino  
PDFpdf ATL-COM-DAQ-2020-110-fig4.pdf r1 manage 28.5 K 2021-03-30 - 04:03 MasayaIshino  
PNGpng ATL-COM-DAQ-2020-110-fig4.png r1 manage 106.9 K 2021-03-30 - 04:06 MasayaIshino  
PDFpdf ATL-COM-DAQ-2020-110-fig5.pdf r1 manage 28.5 K 2021-03-30 - 04:03 MasayaIshino  
PNGpng ATL-COM-DAQ-2020-110-fig5.png r1 manage 99.6 K 2021-03-30 - 04:06 MasayaIshino  
PDFpdf ATL-COM-DAQ-2020-110-fig6.pdf r1 manage 26.9 K 2021-03-30 - 04:03 MasayaIshino  
PNGpng ATL-COM-DAQ-2020-110-fig6.png r1 manage 98.0 K 2021-03-30 - 04:06 MasayaIshino  
Unknown file formateps ATL-COM-DAQ-2022-070-fig1.eps r1 manage 14.6 K 2022-08-26 - 03:38 YukiMitsumori  
PDFpdf ATL-COM-DAQ-2022-070-fig1.pdf r1 manage 14.5 K 2022-08-26 - 03:38 YukiMitsumori  
PNGpng ATL-COM-DAQ-2022-070-fig1.png r1 manage 19.1 K 2022-08-26 - 03:38 YukiMitsumori  
Unknown file formateps ATL-COM-DAQ-2022-070-fig2.eps r1 manage 11.6 K 2022-08-26 - 03:38 YukiMitsumori  
PDFpdf ATL-COM-DAQ-2022-070-fig2.pdf r1 manage 14.6 K 2022-08-26 - 03:38 YukiMitsumori  
PNGpng ATL-COM-DAQ-2022-070-fig2.png r1 manage 15.1 K 2022-08-26 - 03:38 YukiMitsumori  
Unknown file formateps ATL-COM-DAQ-2022-070-fig3.eps r1 manage 12.7 K 2022-08-26 - 03:38 YukiMitsumori  
PDFpdf ATL-COM-DAQ-2022-070-fig3.pdf r1 manage 14.9 K 2022-08-26 - 03:38 YukiMitsumori  
PNGpng ATL-COM-DAQ-2022-070-fig3.png r1 manage 15.1 K 2022-08-26 - 03:38 YukiMitsumori  
Unknown file formateps ATL-COM-DAQ-2022-070-fig4.eps r1 manage 13.4 K 2022-08-26 - 03:48 YukiMitsumori  
PDFpdf ATL-COM-DAQ-2022-070-fig4.pdf r1 manage 14.9 K 2022-08-26 - 03:48 YukiMitsumori  
PNGpng ATL-COM-DAQ-2022-070-fig4.png r1 manage 16.0 K 2022-08-26 - 03:48 YukiMitsumori  
Edit | Attach | Watch | Print version | History: r12 < r11 < r10 < r9 < r8 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r12 - 2022-08-26 - MasayaIshino
 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Atlas All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright &© 2008-2023 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback