TRIDAQ Systems in HEP Experiments at LHC Accelerator

Abstract

TRIDAQ Systems in HEP Experiments at LHC Accelerator, IJET 2013, 59(4), 415-421;

The paper describes Trigger and Data Acquisition (TRIDAQ) systems of accelerator experiments for High Energy Physics. The background for physics research comprises assumptions of the Standard Model theory with basic extensions. On this basis, a structure of particle detector system is described, with emphasis on the following functional blocks: Front-End Electronics, Trigger and DAQ systems. The described solutions are used in the LHC experiments: ATLAS, ALICE, CMS and LHCb. They are also used in other accelerator experiments. Data storage and processing functionality is divided into two hardware systems: Trigger and Data Acquisition, that are dependent on each other. High input data rate impose relevant choices for the architecture and parameters of both systems. The key parameters include detailed system structure and its overall latency. Trigger structure is defined by the physics requirements and the storage capability of DAQ system. Both systems are designed to achieve the highest possible space and time resolution for particle detection. Trigger references are reviewed [1]-[39] as well as chosen accelerator research efforts originating in this country [40]-[83].

Full text

INTL JOURNAL OF ELECTRONICS AND TELECOMMUNICATIONS, 2013, VOL. 59, NO. 4, PP. 415–421
Manuscript received March 18, 2013; revised October, 2013. DOI: 10.2478/eletel-2013-0051
TRIDAQ Systems in HEP Experiments
at LHC Accelerator
Agnieszka Zago´zdzi´
nska, Ryszard S. Romaniuk, Krzysztof T. Po´zniak, and Piotr Zalewski
Abstract—The paper describes Trigger and Data Acquisition
(TRIDAQ) systems of accelerator experiments for High Energy
Physics. The background for physics research comprises assump-
tions of the Standard Model theory with basic extensions. On
this basis, a structure of particle detector system is described,
with emphasis on the following functional blocks: Front-End
Electronics, Trigger and DAQ systems. The described solutions
are used in the LHC experiments: ATLAS, ALICE, CMS and
LHCb. They are also used in other accelerator experiments. Data
storage and processing functionality is divided into two hardware
systems: Trigger and Data Acquisition, that are dependent on
each other. High input data rate impose relevant choices for the
architecture and parameters of both systems. The key parameters
include detailed system structure and its overall latency. Trigger
structure is defined by the physics requirements and the storage
capability of DAQ system. Both systems are designed to achieve
the highest possible space and time resolution for particle
detection. Trigger references are reviewed [1]–[39] as well as
chosen accelerator research efforts originating in this country
[40]–[83].
Keywords—TRIDAQ, trigger, RPC, CMS, HEP, electronic
systems, detectors, front-end electronics, CERN, data acquisition
systems, distributed measuring and control networks, photonic
networks
I. INT ROD UC TI ON
HIGH ENERGY PHYSICS (HEP) is aimed at research of
fundamental particles and forces to explain the structure
of matter. The Standard Model (SM) [1], [2] theory describes
phenomena up to the scale of 10−20 m. It is smaller than
the scale of electroweak forces which are revealed at the
distance range of 10−18 m to equalize with the EM fields.
The SM groups fermions in generations, like the symmetry
group SU(3)cSU(2)×U(1) and assumes that particle masses
are generated by the Higgs mechanism. There are some
suggestions that the SM must be extended to contain newly
discovered mechanisms. SM theory does not contain gravity
forces that are comparable to electromagnetic for energy level
greater than 100GeV, via the masses of W+, W−and Z◦
bosons. Without a new and more fundamental theory, most
of questions cannot be resolved.
There are three main directions of theoretical speculations:
technicolors models, supersymmetry theories and additional
dimensions. Each of them differently explains a problem of
fitting and unification of different kind of forces. Technicolor
models [3] suggest that the Higgs boson is not a point particle
This work was supported in part by Polish Ministry of Science and Higher
Education grant no. N N202 167440.
A. Zago´zdzi´
nska, R. S. Romaniuk, and K. T. Po´zniak are with Warsaw
Univ. of Technology, and P. Zalewski is with National Center for Nuclear
Research (e-mail: azago@cern.ch).
but has an internal structure that will be discovered at energy
1TeV. This theory is probably impossible to be mapped into
realistic models. Supersymetric theory [4] assumes that each
particle has a symmetrical shadow partner which spin value
differs of 1/2. It implies that each fermion with 1/2 spin
has its pair in boson with spin value 0. Their names are
created by adding s- prefix like selectron or squark. Each
0 or 1 spin boson has the corresponding fermion with spin
value 1/2. Their names are created by adding sufixes -ino like
gluino or higgsino. The supersymetric family contains double
number of fundamental particles with masses above 1TeV. It is
compatible with the Great Unification Theory (GUT) that es-
tablishes equalization of electroweak and electrostrong forces
at the energy level about 1016 GeV. The GUT theory expands
the SM but has not been experimentally acknowledged yet.
The newest concept contains additional dimensions [5]. For
electroweak forces up to 10−18 m scale spacetime is certainly
four-dimensional. The previous considerations ignored grav-
ity forces for interaction scale lower than 1mm. Additional
dimensions for the gravity theory change the growth rate
of gravitational constant. That results in strong influence of
the gravity above the energy value of 1TeV. Influence of the
gravity allows to solve the fine tuning problem [6].
Current and planned future accelerator and non accelerator
experiments are supposed to verify these theories or prove that
SM is a fundamental theory and cannot be developed further.
To explore interactions of scale lower than 10−18 m, special-
ized detectors for the particle research are needed [7]. Most
of them contain multilevel triggering system and expanded
data acquisition system but differ by the used technology [8].
Detectors must be sensitive to the presence of the particles
having different charge, that are predicted by theory. It is
possible by generating strong electromagnetic fields where
the particles flight path is curved and registered. Electronic
systems that trigger data registration must be fast, reliable and
resistant to high radiation levels. Non accelerator experiments
(Auger Project [9], DAMA [10]) explore particles produced
in natural processes. Most of such detectors are hidden deep
underground and are aimed to register extremely rare particles.
This paper is focused on the trigger and data acquisition
systems for the detector experiments in Large Hardron Collider
(LHC) [11] at CERN.
II. DE TE CT OR SY ST EM ST RUC TU RE
Experiments requirements, budget and current technical
capabilities define the detectors size and structure. ATLAS [12]
and CMS [13] detectors in LHC are designed to investigate
Unauthenticated
Download Date | 1/14/16 9:22 AM

416 A. ZAGO ´
ZDZI ´
NSKA, R. S. ROMANIUK, K. T. PO´
ZNIAK, P. ZALEWSKI
Fig. 1. Example of the data distribution produced in the different CMS
sub-systems for proton-proton collisions.
the products of particle collisions. Both are general purpose
detectors that have recently discovered the Higgs boson,
and are expected to discover extra dimensions, as well as
particles responsible for dark matter existence. Successive
layers in the detectors are: the tracker, EM calorimeters,
hadronic calorimeters and muon systems. General structure of
the detectors is similar but all the systems and subsystems have
different internal structure. The most visible difference is in the
detectors size. ATLAS weight is only 7000 ton and the volume
is 20 000 m3while the CMS is 12 500 ton and about 3800
m3. Detector technology is not the only difference between
the experiments. Collision rate and the rate of registered data
determines the number of channels and their density. In LHCb
detector [14], [15], which is a stationary target, the luminosity
is low and only one side of the collision is investigated. The
number of channels is lower than in most of the detectors in
LHC, but the amount of registered data is higher. The input
rate to HLT (high level trigger) is about 1MHz in LHCb, while
for CMS it is reduced to 100kHz.
This task is performed by a multilevel triggering and data
acquisition system realized using fast, custom made electron-
ics. Much slower, high level trigger system, implemented in
computer farms, receives a reduced amount of the data. The
input rate is different in each part of the detector and sub-
detectors. In CMS 75% from 100 million electronic channels
comes from the tracker sub-detector. The particles are detected
by the silicon strips and pixels. The pixel readouts supply 6000
connections per square centimeter, that is 65 million pixels in
the whole sub-detector. 65% of the output signals is produced
by the pixel readout in the tracker sub-detector. The average
event sizes, for different CMS subsystems, for p-p collisions,
are shown in Fig. 1. Each shade of grey corresponds to the
separate subsystem. Some detector chambers transmit more
data than the others. That is marked as the length difference
of some wedges. The whole system is designed for an event
size of 1 MB, at the maximum luminosity.
III. FRON T END EL EC TRO NI CS
Signals from the detector are collected by the front-end
boards. The electronics connected directly to the detector
must be sensitive to short current pulses and resistant to
high level radiation. The detector response depends on the
charge deposit, event rate and sensitiveness of the device. Its
measurement precision is limited by the charge absorption in
the detector and electronic noise. Signal charge for a given
energy absorption is formed by many elementary excitations
[18]. Increase of the measurement resolution improves the
signal to background ratio but limits the maximum signal
bandwidth. For this reason, fast rising signals are shaped as
long as there is a spacing between the pulses. This can be
difficult to obtain when the time interval between the collisions
is low. If the beam is continuous, the distinction between
bunch crossings is possible only with the time measurement.
Most of the detectors have particular drift time that limits
the time resolution. For Drift-tubes detectors the minimum
time resolution is 1ns, because location of the particle cannot
be accurately determined. Better resolution is achieved in the
Time Projection Chambers. Very high time resolution, up to
10-100ps, can be achieved only in the Time Of Flight (TOF)
detectors. Such detectors are used e.g. in ALICE experiment
[19]. The measurement resolution is also limited by the time
dependency on amplitude. Its influence can be reduced by the
use of constant fraction discriminators. The main disadvantage
of this solution is the additional time delay. The amplitude
can be compensated by the use of two channels of a Time to
Digital Converter (TDC). Each of them is set for a different
threshold [20]. In systems of the size comparable to ATLAS or
CMS, the number of such modules is a few thousands. Their
duplication would significantly increase the measurement time
and cost. The alternative method is Time Over Threshold
(TOT), where both leading edge and pulse width are measured
in each channel. Such system is used in the Atlas Pixel
Detector [21].
The amplified and shaped analog signals are converted to
digital. This conversion process can be executed in various
places of the processing chain. Depending on the conversion
site, the systems can be divided into the following categories:
analog readout and buffering, digital readout with analog
buffering or both digital. In majority of the systems, the
conversion is realized as soon as possible. Converters must
provide sufficiently high level of the sensitivity and accuracy
within a given budget. High resolution converters must work
with low jitter clocks to avoid the additional source of noise.
If the A/D converters have an insufficient speed or their
power consumption is too high, analog buffers are used. Such
buffers are used both in small storage systems, like sampling
oscilloscopes, and in complex multichannel systems like the
CMS tracker, Atlas calorimeter or LHCb trackers. Even if
the buffers decrease the output rate, it can be still too high
for what the Data Acquisition (DAQ) system can handle. The
real life data rate cannot be estimated before the experiment is
Unauthenticated
Download Date | 1/14/16 9:22 AM

TRIDAQ SYSTEMS IN HEP EXPERIMENTS AT LHC ACCELERATOR 417
Fig. 2. Inclusive proton-(anti)proton cross sections σfor basic physics
processes. Interaction rates for the nominal luminosity are given on the
righthand scale.
running. The estimation is based on physical simulations and
experiences of the previous experiments. This can cause some
problems in the hardware design. The buffers depth is limited
by the cost and technical capabilities. If these parameters
are too small there is a risk of the data overflows. In many
systems the zero-suppression algorithm is implemented. That
enables reduction of the data amount before it is sent to the
trigger. Only non-zero data are transmitted to the readout
system, synchronically to the global bunch-crossing clock.
This method is regarded as the most efficient solution for the
very high-rate future experiments such as SLHC [22] or CLIC
[23].
IV. THE TR IG GE RI NG SY ST EM
The trigger system analyzes events in regular intervals, in
the collider experiments. The particles collide and interact
during each bunch-crossing (BX). The method of the data
selection is defined by the requirements of particular HEP
experiment. The smaller granularity of the matter, the greater
energy, luminosity and collision rate are required for the
research. The rate was only 45kHz for LEP [24], 2.5/7.6MHz
for Tevatron [25] and 40MHz for LHC. For high luminosity
hadron colliders, each BX can provide up to 25 events that
contain different number of muons, clusters etc. This rate
could not be handled by any data acquisition system. The
collision cross section σunit is the barn [b]. The barn describes
the probability of interaction between particles in collider
experiments for HEP. At LHC, the total proton-proton cross
section is about 70mb [26], [27]. For a nominal collision
energy, the frequency of particles generation (Fig. 2) is 70Hz
for Beauty quark (0.7 mb), 1kHz for W/Z bosons (200/60
nb), 8Hz for Top quark (0.8 nb) and 0.3Hz for Higgs 150 GeV
boson (30pb). This means that only small part of the generated
data is useful for the analysis. To cut off an uninteresting
physics information, the multilevel trigger system is used. It
is realized with the use of signatures that contain parameters
such as a threshold, amplitude or more complex data. The
set of parameters is initially based on software simulation and
calculation of muon tracks, energy deposits in the calorimeters,
and track in the silicon detectors. Particles such as muons and
electrons have clear signatures. A separation of a single lepton
from the jets requires an analysis of particle showers. The
trigger systems at the collider experiments are sensitive to the
particles transverse momentum (pT).
The trigger system must be simple and selective, to follow
the incoming data in the real time. The low level triggers
are implemented using specialized electronics with fast FPGA
chips [28]. The high level trigger is realized off line by
computer farms. The final output rate is limited by an offline
computing budget and storage capacity of the system. A dead-
time is determined as a ratio between a time when the data
acquisition system is busy and a total time. An important factor
increasing the dead-time value is random distribution of the
data. Most of the events occur within a short time window,
then the fraction of the system busy time is much higher.
Typically, the system is busy during the ADC conversion and
storage time. In the system enhanced with an additional FIFO
queue, the dead-time is reduced to the conversion time. The
storage element must be large enough to follow the full data
size during the system operating time. The effective trigger rate
is reduced to a level that can be handled by the readout system.
At the CDF Trigger System in Fermilab [29], the average dead-
time is kept below 5% [30]. At the BNL-E949 experiment in
the Alternating Gradient Synchrotron for the Level-0 trigger,
the online dead time was reduced from 4.0% to 1.7% [31].
At the CMS Level-1 trigger, for the maximum output rate of
100kHz, the dead time is estimated to be below 1% [32].
Most of the HEP experiments contain a multilevel trigger
system [33]. The triggering task can be divided to 2 (CMS),
3 (ATLAS and LHCb) or 4 levels (ALICE). The Level-0
Trigger is based on the hardware implementation with the use
of FPGAs. The data cannot be processed longer than a few µs,
so selection algorithms have to be as simple as possible. The
selection is aimed primarily at the identification of leptons.
Analysis covers the low precision data from detectors like
hadron and electromagnetic calorimeters for electrons/γ/jets
or the muon chambers. Its results depend on the values
of programmable parameters such as pTthreshold and BX
identification methods that are implemented. The high level
trigger (HLT) performs more complex algorithms and can be
busy for a few ms up to 10s. The algorithms are implemented
in a software and hardware. The HLT is a selective process.
At some experiments, its functionality can be divided into
two levels: accessing to the part of the event (Level-1) and to
the full event (Level-2). Both parts analyze the full precision
Unauthenticated
Download Date | 1/14/16 9:22 AM

418 A. ZAGO ´
ZDZI ´
NSKA, R. S. ROMANIUK, K. T. PO´
ZNIAK, P. ZALEWSKI
Fig. 3. RPC Trigger efficiency measured by the Tag & Probe method for
the barrel, Data: 2011B, probes with pT> 20GeV from Z’s, probes with pT
< 20GeV from J/ψ’s [36].
and granularity information from the chambers. The HLT uses
the topological variables and tracking information from inner
detectors e.g. the ATLAS calorimeter trigger provides the
information about the particles e, γ,τ, jets and also about
ETmiss,ΣET[34]. The triggering conditions can be imposed
to reduce the jet background at low energy thresholds.
The difficulty for the trigger system is to decrease the
data rate and to enable collection and data processing that
are of interest for physics. Both aims cannot be achieved
simultaneously. When the efficiency grows, the rejection of
the physical background is worse and vice versa, the improved
background rejection means a decrease of the efficiency. In
contrast to the precision experiments, where the well known
selections are used, the discovery experiments use the inclusive
selections. In both cases the rejected data are lost and cannot
be restored. Current efficiency achieved at the RPC PAC
trigger at the CMS is about 92% in the barrel (Fig. 3) and 80%
in the endcaps for pTgreater than 20 GeV/c. Maintenance
of high efficiency with increasing LHC luminosity requires
a continuous work on the trigger patterns optimization. The
system efficiency is defined as a number of events that passed
the selection ratio to a number of all events. There are pass-
through triggers implemented in the HLTs, to select samples
without any bias. At the L0 Trigger, the number of events
without the selection is an unknown parameter. There are
commonly two methods used to estimate the efficiency. The
first one is a comparison between the orthogonal triggers. The
second method is the experimental technique called “Tag-
and-Probe” [35]. In this method, a particle track is selected
(Tag) and a complementary offline track is found (Probe).
The trigger selection is applied on the probe. The capability
of the rate is not unlimited. Some particles cannot reach the
Fig. 4. Components of the DAQ system of CMS [38].
threshold level because their pTis underestimated. The error,
in estimating δpT/pT, can reach 30% for the L0 Trigger, which
has the worst spatial resolution.
V. DATA ACQU ISITION SYSTEM
The detector system divides its functionality between the
trigger and data acquisition systems. A dual architecture
enables separation of the processing and storage functions.
The trigger and DAQ parameters are shown in Tab. I. The
data are buffered in the DAQ system. Data rate is decreased in
the subsequent trigger levels from MHz to kHz. Both systems
work in a real-time mode where the latency is low and strictly
determined. The DAQ system also builds and stores complete
events from a segmented information from the detector. This
process is called an event building and data logging. There
are control, configuration and monitoring tools implemented
in the DAQ. The basic parameters are, similarly to the Trigger
System, the dead time and efficiency.
The DAQ system can be divided into four functional stages:
•a detector readout – collects the data in about 700 buffers
(ATLAS, CMS),
•an event building stage – collects all the data correspond-
ing to a single event,
•a selection stage – implements the HLT functionality in
a processor farm,
•an analysis and storage – selected events are forwarded
to the computing services for storage and analysis.
The general structure of the DAQ system at the CMS
experiment is shown in Fig. 4. Two complementary systems
enable the data flow from Detector Front Ends to processor
farm. The Front-End Electronics is the first storage element
in the system connected directly to the detecting units of each
sub detector. There are approximately 700 modules in the
CMS readout [39]. The data from Front-End Boards (FEB) are
read and stored in the Readout System buffers. The storage
is realized until the data can be sent further to a processor
Unauthenticated
Download Date | 1/14/16 9:22 AM

TRIDAQ SYSTEMS IN HEP EXPERIMENTS AT LHC ACCELERATOR 419
TABLE I
TRI GG ER AN D DAQ PA RA MET ER S [37]
Detector T rigg er Rate[Hz ]Event siz e Readout HLT O ut
levels [Byte] [GB/s] [M B/s] (E vents/s)
ALICE 4 L3 Pb-Pb 500 5×1072 1250 (102)
L3 p-p 1032×106200 (102)
ATLAS 3 L0 1051,5×1064,5 300 (2×102)
L1 3×103
L2 2×102
CMS 2 L0 105106100 ∼1000 (102)
L1 102
LHCb 2 L0 1063,5×10435 70 (2×103)
L1 4×103
that analyze the event. The buffers are deep enough to wait
until the decision from the L1 trigger is not ready. The
Readout Unit supports the number of the FEBs. There are
about 500 such Readout Columns. A positive trigger signal
enables the further data transmission to the HLT processor
farm. The transmission is through the Builder Network which
is a large switch network with the 800Gb/s throughput. To
provide the continuous data flow without traffic jams, the
control signals are also sent through the switch. The Filter
Systems process the events to realize the HLT algorithms. The
selected events are finally stored for the offline processing.
There are about 500 Filter Columns in the CMS. Each of
them contains one Builder Unit and multiple Filter Units. The
Builder creates a complete single event from the incoming
data fragments. The single Filter Unit contains processors for
the HLT algorithms. The filtered events, and small fraction
of the rejected events, are sent to the Computing Services.
The Event Manager is a unit responsible for the DAQ data
flow. It centralizes the event management and synchronizes
the overall system. A Control and Monitor unit is aimed to
configure, control and monitor all system elements. It manages
the interface to the DAQ offline environment. Such modular
architecture of the DAQ system and its division into the four
main tasks enables independent implementation and testing. It
also requires a usage of an additional deep buffer. The buffers
enable matching between all the operating stages, working at
different rates.
VI. CO NCLUSION AND SUMMARY
The TRIDAQ systems must be consistent with the physics
goals. At the LHC experiments the Trigger Systems are
aimed to follow the rate of 200kHz. The limitations are: the
bandwidth, required efficiency and background rejection. The
trigger rate during the design and commissioning is based
on simulations containing large amount of the data with
cross-section background. Initial analysis must include the
information about the detector topology. The trigger must be
flexible enough to cope with the real conditions that could be
different than expected. The system must provide a channel
masking to reject the channels that give much higher rate
than expected. A typical solution is a programmable threshold
implementation. The threshold level reduction increases the
number of registered events. That is desirable until profits are
higher than the offline computing cost. The data are analyzed
simultaneously by the different triggering algorithms to follow
the physics requirements. The trigger menu contains a set of
parameters for each level of the trigger – L0, L1 and L2,
that are grouped vertically to create a trigger chains. In an
ideal situation, the triggered and stored data should overlap to
save the storage space. There are two types of the TRIDAQ
systems: dedicated to collect the data from accelerator and
non accelerator experiments. The Trigger Systems for the
HEP experiments are independent but complementary to the
DAQ systems. Both are built with smaller blocks that can be
designed and implemented separately. This is an important
feature when the project is realized by groups from institutes
scattered around the world. A specification of these systems
is determined by the HEP predictions and experiences of the
previous experiments. There is a clear trend of increasing
a number of the collected data. The growth is a result of
a few factors such as the increase of energy at accelerator
experiments or the newest achievements in the magnets area.
These factors enable better beam positioning and focusing that
influence the reached luminosity and the number of triggered
data. The current systems are designed to achieve the highest
space and time resolution. The multiplication of the input
channels has also an influence on the registered data size. All
these factors seems to exist in the current and the nearest future
experiments that are challenging to the TRIDAQ systems. This
growth could be stopped if the registered event resolution
exceeds the analysis and acquisition capability.
This work was presented during the Wilga 2012 Symposium
on Electronic and Photonic Systems for High Energy Physics
Experiments and Astronomy and in part published in the Wilga
2012 proceedings.
REF ER EN CE S
[1] M. Schmaltz, “Physics beyond the standard model (Theory): Introducing
the Little Higgs,” Nuclear Physics B – Proceedings Supplements, vol.
117, pp. 40–49, 2003.
[2] C. Quigg, “The Standard Model (Electroweak Theory),” European
School of High-Energy Physics, 2002.
[3] J. Andersen et al., “Discovering technicolor,” The European Physical
Journal Plus, vol. 126, no. 81, 2011.
Unauthenticated
Download Date | 1/14/16 9:22 AM

420 A. ZAGO ´
ZDZI ´
NSKA, R. S. ROMANIUK, K. T. PO´
ZNIAK, P. ZALEWSKI
[4] P. Binétruy, Supersymmetry: Theory, Experiment, and Cosmology. Ox-
ford University Press, 2006.
[5] Website http://cosmology.berkeley.edu/inpac/CDMSCE_Jun06/
Talks/ cembranos_adddimensions.pdf.
[6] R. Kitano and Y. Nomura, “A Solution to the Supersymmetric Fine-
Tuning Problem within the MSSM,” Physics Letters B, vol. 631, pp.
58–67, 2005.
[7] A. Zago ´zdzi´
nska, K. T. Po´zniak, and R. Romaniuk, “Heavy stable
charged particles search by novel pattern comparator processor,” Pro-
ceedings of SPIE, vol. 8454, 2012.
[8] K. T. Po´zniak, “Electronics and photonics for high energy physics
experiments,” Proceedings of SPIE, vol. 5125, 2002.
[9] Website http://www.auger.org/.
[10] Website http://people.roma2.infn.it/ dama/web/home.html.
[11] Website http://lhc.web.cern.ch/lhc/.
[12] Website http://atlas.web.cern.ch/Atlas/Collaboration/.
[13] Website http://cms.web.cern.ch/.
[14] Website http://lhcb.web.cern.ch/lhcb.
[15] The LHCb Collaboration, “The LHCb Detector at the LHC,” Journal of
Instrumentation, vol. 3, 2008.
[16] D. Froidevaux and P. Sphicas, “General-purpose detectors for the Large
Hadron Collider,” Annual Review of Nuclear and Particle Science,
vol. 56, pp. 375–440, 2006.
[17] The CMS Collaboration, “The data-acquisition system of the CMS
experiment at the LHC,” Journal of Physics: Conference Series 331,
2011, 022021.
[18] H. Spieler, “Front-End Electronics and Signal Processing,” in Instru-
mentation in Elementary Particle Physics, AIP Conference Proceedings,
2002, vol. 674.
[19] Website http://aliceinfo.cern.ch/Public/en/Chapter2/Chap2_TOF.html.
[20] J. Christiansen, “HPTDC High Performance Time to Digital Converter,”
HPTDC, vol. 1.3, 2004.
[21] Website http://hepwww.rl.ac.uk/Vertex03/Talks/JornGrosseKnetter.pdf.
[22] Website http://project-slhc.web.cern.ch/project-slhc/about/.
[23] Website http://clic-study.org/.
[24] Website http://public.web.cern.ch/public/en/Research/LEPExp-en.html.
[25] Website http://www-bdnew.fnal.gov/tevatron/.
[26] The CMS Collaboration, “CMS The TriDAS ProjectTechnical Design
Report, Vol.1: The Trigger Systems,” CERN/LHCC 2000 – 38 CMS
TDR 6.1, 2000.
[27] G. Bayatian, S. Chatrchyan, J. Krolikowski, K. Po´zniak, and
W. Zabołotny et al., “CMS physics technical design report, volume
II: Physics performance,” Journal of Physics G: Nuclear and Particle
Physics, vol. 34, no. 6, 2007, index: S01.
[28] K. T. Po´zniak, “FPGA-based, specialized trigger and data acquisition
systems for high-energy physics experiments,” Measurement Science and
Technology, vol. 21, no. 6, 2010, index: 62002.
[29] Website http://www-cdf.fnal.gov/.
[30] The CDF Collaboration, “Using Feedback to Control Deadtime in the
CDF Trigger System,” 2010.
[31] T. Yoshioka et al., “Upgrade of the Level-0 Trigger System for BNL-
E949,” IEEE Transactions on Nuclear Science, vol. 51, pp. 334–339,
2004, issue 3.
[32] I. Magrans de Abril, “The CMS Trigger Supervisor: Control and
Hardware Monitoring System of the CMS Level-1 Trigger at CERN,”
2008.
[33] M. Mozer, “Triggers for New Physics at the LHC,” Journal of Physics:
Conference Series 171 012101, 2009.
[34] R. Achenbach et al., “The ATLAS Level-1 Calorimeter Trigger,” JINST,
vol. 3, 2008, P03001.
[35] The ATLAS Collaboration, “Tag and Probe method for Electron ID
efciencies,” 2009.
[36] M. Konecki, Performance of the RPC trigger. CMS Warsaw Group,
2012.
[37] N. Neufeld, Status and Prospects of LHC Experiments Data Acquisition.
CERN/PH, 2009.
[38] The CMS Collaboration, “CMS The TriDAS Project Technical De-
sign Report, Volume 2: Data Acquisition and High-Level Trigger,”
CERN/LHCC 02-26 CMS TDR 6, 2002.
[39] A. Zago´zdzi´
nska, R. S. Romaniuk, K. T. Po´
nniak, and P. Zalewski,
“Tridaq systems in HEP experiments at LHC accelerator,” Proceedings
of SPIE, in press, 2013.
[40] R. S. Romaniuk, “Technika akceleratorowa i eksperymenty fizyki
wysokich energii,” Wilga 2012, Elektronika – konstrukcje, technologie,
zastosowania, vol. 53, no. 9, pp. 162–169, 2012.
[41] R. S. Romaniuk, et al., “Optical network and FPGA/DSP based control
system for free electron laser,” Bulletin of the Polish Academy of
Sciences, Technical Sciences, vol. 53, no. 2, pp. 123–138, 2005.
[42] R. S. Romaniuk, “Polfel – a free electron laser in Poland,” Photonics
Letters of Poland, vol. 1, no. 3, pp. 103–105, 2009.
[43] B. Mukherjee, et al., “Application of low-cost GaAs LEDs as kerma
dosemeters and fluence monitor for high-energy neutrons,” Radiation
Protection Dosimetry, vol. 126, no. 1–4, pp. 256–260, 2007.
[44] R. Romaniuk, et al., “Metrological aspects of accelerator technology and
high energy physics experiments,” Measurement Science and Technol-
ogy, vol. 18, no. 8, 2008, art.no.E01.
[45] P. Fafara, et al., “FPGA-based implementation of a cavity field controller
for FLASH and X-FEL,” Measurement Science and Technology, vol. 18,
no. 8, pp. 2365–2371, 2008.
[46] W. Ackerman, et al., “Operation of a free-electron laser from the extreme
ultraviolet to the water window,” Nature Photonics, vol. 1, no. 6, pp.
336–342, 2007.
[47] T. Czarski, et al., “Superconducting cavity driving with FPGA con-
troller,” Nuclear Instruments and Methods in Physics Research A, vol.
568, no. 2, pp. 854–862, 2006.
[48] ——, “TESLA cavity modeling and digital implementation in FPGA
technology for control system development,” Nuclear Instruments and
Methods in Physics Research A, vol. 556, no. 2, pp. 565–576, 2006.
[49] ——, “Cavity parameters identification for TESLA control system
development,” Nuclear Instruments and Methods in Physics Research
A, vol. 548, no. 3, pp. 283–297, 2005.
[50] R. Romaniuk, “Accelerator infrastructure in Europe EuCARD 2011,”
International Journal of Electronics and Telecommunications, vol. 57,
no. 3, pp. 413–419, 2011.
[51] ——, “EuCARD 2010 accelerator technology in Europe,” International
Journal of Electronics and Telecommunications, vol. 56, no. 4, pp. 485–
488, 2010.
[52] S. Chatrchyan, et al., “The CMS experiment at the CERN LHC,” Journal
of Instrumentation, vol. 3, no. 8, 2008, art.no S08004.
[53] ——, “Commissioning of the CMS experiment and the cosmic run
at four tesla,” Journal of Instrumentation, vol. 5, no. 3, 2010,
art.no.T03001.
[54] ——, “Performance of the CMS Level-1 trigger during commissioning
with cosmic ray muons and LHC beams,” Journal of Instrumentation,
vol. 5, no. 3, 2010, art.no.T03002.
[55] ——, “Performance of the CMS drift-tube chamber local trigger
with cosmic rays,” Journal of Instrumentation, vol. 5, no. 3, 2010,
art.no.T03003.
[56] ——, “Fine synchronization of the CMS muon drift-tube local trigger
using cosmic rays,” Journal of Instrumentation, vol. 5, no. 3, 2010,
art.no.T03004.
[57] ——, “Commissioning of the CMS High-Level Trigger with cosmic
muons,” Journal of Instrumentation, vol. 5, no. 3, 2010, art.no.T03005.
[58] ——, “CMS data processing workflows during an extended cosmic ray
run,” Journal of Instrumentation, vol. 5, no. 3, 2010, art.no.T03006.
[59] ——, “Commissioning and performance of the CMS pixel tracker with
cosmic ray muons,” Journal of Instrumentation, vol. 5, no. 3, 2010,
art.no.T03007.
[60] ——, “Measurement of the muon stopping power in lead tungstate,”
Journal of Instrumentation, vol. 5, no. 3, 2010, art.no.P03007.
[61] ——, “Comissioning and performance of the CMS silicon strip tracker
with cosmic ray muons,” Journal of Instrumentation, vol. 5, no. 3, 2010,
art.no.T03008.
[62] ——, “Alignment of the CMS silicon tracker during commissioning
with cosmic rays,” Journal of Instrumentation, vol. 5, no. 3, 2010,
art.no.T03009.
[63] ——, “Performance and operation of the CMS electromagnetic
calorimeter,” Journal of Instrumentation, vol. 5, no. 3, 2010,
art.no.T03010.
[64] ——, “Time reconstruction and performance of the CMS electromag-
netic calorimeter,” Journal of Instrumentation, vol. 5, no. 3, 2010,
art.no.T03011.
[65] ——, “Performance of the CMS hadron calorimeter with cosmic ray
muons and LHC beam data,” Journal of Instrumentation, vol. 5, no. 3,
2010, art.no.T03012.
[66] ——, “Performance of CMS hadron calorimeter timing and synchro-
nization using test beam, cosmic ray, and LHC beam data,” Journal of
Instrumentation, vol. 5, no. 3, 2010, art.no.T03013.
[67] ——, “Identification and filtering of uncharacteristic noise in the CMS
hadron calorimeter,” Journal of Instrumentation, vol. 5, no. 3, 2010,
art.no.T03014.
Unauthenticated
Download Date | 1/14/16 9:22 AM

TRIDAQ SYSTEMS IN HEP EXPERIMENTS AT LHC ACCELERATOR 421
[68] ——, “Performance of the CMS drift tube chambers with cosmic rays,”
Journal of Instrumentation, vol. 5, no. 3, 2010, art.no.T03015.
[69] ——, “Calibration of the CMS drift tube chambers and measurement of
the drift velocity with cosmic rays,” Journal of Instrumentation, vol. 5,
no. 3, 2010, art.no.T03016.
[70] ——, “Performance study of the CMS barrel resistive plate chambers
with cosmic rays,” Journal of Instrumentation, vol. 5, no. 3, 2010,
art.no.T03017.
[71] ——, “Performance of the CMS cathode strip chambers with cosmic
rays,” Journal of Instrumentation, vol. 5, no. 3, 2010, art.no.T03018.
[72] ——, “Aligning the CMS muon chambers with the muon alignment
system during an extended cosmic ray run,” Journal of Instrumentation,
vol. 5, no. 3, 2010, art.no.T03019.
[73] ——, “Alignment of the CMS muon system with cosmic-ray and
beam-halo muons,” Journal of Instrumentation, vol. 5, no. 3, 2010,
art.no.T03020.
[74] ——, “Precise mapping of the magnetic field in the CMS barrel yoke
using cosmic rays,” Journal of Instrumentation, vol. 5, no. 3, 2010,
art.no.T03021.
[75] ——, “Performance of CMS muon reconstruction in cosmic-ray events,”
Journal of Instrumentation, vol. 5, no. 3, 2010, art.no.T03022.
[76] R. S. Romaniuk, “Accelerator Technology and High Energy Physics
Experiments, Photonics Applications and Web Engineering, Wilga May
2012,” Proceedings of SPIE, vol. 8454, 2012, art.no. 845403.
[77] ——, “Physics and Plasma Research, Photonics Applications and Web
Engineering, Wilga May 2012,” Proceedings of SPIE, vol. 8454, 2012,
art.no. 845404.
[78] ——, “Accelerator Science and Technology in Europe – EuCARD
2012,” Proceedings of SPIE, vol. 8454, 2012, art.no. 845407.
[79] ——, “Proceedings of SPIE,” 2011, art.no. 800802.
[80] ——, “Accelerator infrastructure in Europe: Eucard 2012,” Proceedings
of SPIE, vol. 8008, 2011, art.no. 800805.
[81] ——, “Free electron laser infrastructure in Europe 2012,” Proceedings
of SPIE, vol. 8703, 2013, art.no. 870323.
[82] ——, “Accelerator science and technology in Europe – EuCARD 2012,”
International Journal of Electronics and Telecommunications, vol. 58,
no. 4, pp. 327–334, 2012.
[83] ——, “Space and high energy experiments – Advanced electronic
systems 2012,” International Journal of Electronics and Telecommu-
nications, vol. 58, no. 4, pp. 441–462, 2012.
Unauthenticated
Download Date | 1/14/16 9:22 AM