• Aucun résultat trouvé

LHCb muon trigger

N/A
N/A
Protected

Academic year: 2021

Partager "LHCb muon trigger"

Copied!
17
0
0

Texte intégral

(1)

HAL Id: in2p3-00011395

http://hal.in2p3.fr/in2p3-00011395 Submitted on 28 Mar 2002

HAL is a multi-disciplinary open access

archive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

LHCb muon trigger

A. Tsaregorodtsev

To cite this version:

(2)

2nd International Symposium 28-30 June 2000

LHC Physics and Detectors JINR, Dubna, Russia

LHCb Muon Trigger

A.Tsaregorodtsev1

on behalf of the LHCb Collaboration

1.

Introduction

2.

Algorithm

3.

Performance

4.

Implementation

5.

Conclusion

(3)

2

I

NTRODUCTION

Physics motivation for the muon trigger:

● Selects efficiently channels of interest:

➥ BdJ/ψ(µ+µ−)KS

➥ BsJ/ψ(µ+µ−)φ

➥ Bs → µ+µ− , Bs,d → Κ∗0 µ+µ−

➥ ...

● Selects generic B→µX events

(4)

I

NTRODUCTION

...

● The Muon trigger uses information from all the 5 Muon stations

(5)

I

NTRODUCTION

...

● The Muon trigger uses information from the Muon system. The Muon trigger

identifies muons in the event and measures their transverse momentum PT

200 400 600 0 2 4 6 8 10 0 PT [GeV/c] Muons in Minimum bias events: π’s and K’s

Muons from

B → µX decays

Highest PT muon in Bd → J/ψ(µ+µ−)KS

PT of the muons hitting the Muon system:

(6)

A

LGORITHM

:

MUON TRACK FINDING

Find muon track candidate

➔ Find a seed pad hit in station M3 ➔ Find pads within opened search

windows in stations M2, M4, M5

➔ Use pads found in stations M2 and

M3 to extrapolate to station M1

➔ Find a pad within the opened

search window in station M1

Muon trigger exploits multiple scattering in the matter of the Muon shield by applying tight search windows while muon track finding.

(7)

A

LGORITHM

:

APPLYING

P

T CUT

Calculate PTof the muon candidate from the pad positions in M1 and M2

➔ Approximately equal contributions

of the Muon chambers precision and multiple scattering in the

calorimeters to the PT resolution

(8)

P

ERFORMANCE

...

Efficiency

The Muon trigger ensures Minimum bias event suppression factor 50-100 ➥ 10 - 20% of the L0 total bandwidth

0 0.2 0.4 0.6 0.8 1 0 1 2 3 4 5 6 0 100 200 300 PT [GeV/c] Signal acceptance

Minimum bias suppr

(9)

P

ERFORMANCE

...

Robustness against background

MB ret. 2%, nominal background in M 1

MB ret. 2%, maximal background in M 1

MB ret. 1%, nominal background in M 1 MB ret. 1%, maximal background in M 1

ε = ( 48.8 A 0.8(stat) +0-7(Bckg) +0-HLHC) )Y

ε = ( 31.4 A 1.2(stat) +0-9(Bckg) +0-HLHC) )Y

Background scale factor in M 2-M5

(10)

I

MPLEMENTATION

:

REQUIREMENTS AND SOLUTION

● Muon trigger processor implementation requirements:

➥ minimal efficiency losses with respect to the “theoretical” algorithm ➥ minimal amount of data transfer between the processor and

the detector front-end electronics

➥ fixed latency for taking a decision ➥ easy to monitor and debug

● Specialised hardware processor

➥ two independant processors for the right and left halves of the Muon System ➥ two stages of the data processing

(11)

I

MPLEMENTATION

:

GENERAL ARCHITECTURE

M1 M2 M3 M4 M5

Half Muon Station Front-End Electronics L0 Muon Trigger

12 960 LVDS Links 92 Optical Links (1.6 Gb/s)

Decision Sector Builder Dual Port Memory Fast Identification Processor Detailed Processor 4608 logical channels for each Bunch Crossing

2784 logical channels for each Bunch Crossing

2784 logical channels for each Bunch Crossing

1392 logical channels for each Bunch Crossing

1392 logical channels for each Bunch Crossing

Full Sector Hit Map for each Bunch Crossing

Sector addresses for each FIP candidate

Sector addresses and sector content for each FIP candidate

for each Bunch

Muon

(12)

I

MPLEMENTATION

:

FAST MUON IDENTIFICATION

48 sectors for station M3

192 sector inputs + BCID (M2 to M5)

48 sectors (M3)

48 (M3) 48

48 sectors for station M3

192 sector inputs + BCID (M2 to M5)

48 sectors (M3)

48 (M3) 48

48 sectors for station M3

192 sector inputs + BCID (M2 to M5)

48 sectors (M3) 48 (M3) 48 48 sectors 48 sectors 48 48 1 2 3 4 5 6 7 8 1 2 3 4 5 6 FIP DMP M2 M3 M4 M5

● Each Fast Identification Processor

(FIP) board receives 48×4 sector data from 4 Muon stations M2-M5 each 25 ns

● It finds candidate tracks as a set of

sectors lining up in all the 4 Muon stations M2-M5

● The sector addresses defining a

(13)

I

MPLEMENTATION

:

DETAILED MUON PROCESSING

FIP CONTROLLER DMP CONTROLLER DMP board 1 2 3

...

6 DMP board DMP board DMP board L0 decision Sector addresses Sector contents Synchro check Pad positions to FE from FE

● Realizes theoretical algorithm

in the area selected by the FIP processor

● Asynchronous processing of

track candidates in DMP boards

● Results of the processing are

collected by the DMP controller

● DMP controller restores the

(14)

I

MPLEMENTATION

...

● Latency

● Logistics

➥ 2 VME64 crates on the sides of the Muon System ➥ 32 9U VME64 boards

➥ 184 1.6 Gbit/s optical links

➥ up to 12 optical links on a board

Stage

Time [ns]

FIP Processor 200

ODE interrogation 300

DMP Processor 700

(15)

I

MPLEMENTATION

:

TECHNICAL CHALLENGES

● Large amounts of information to be received and processed in one board

➥ using multiple optical links per board (up to 20)

➔ data synchronization ➔ low power consumption

➥ using circuits of grand integration - very large FPGA’s

● Transfer of large amounts of data inside the processor:

➥ exchange of information between the cards:

point-to-point connection at up to 480 Mhz

➥ collection of data from multiple sources in one destination:

multipoint (up to 16) connection at 40 MHz

● Synchronisation of a large number of distributed devices

➥ synchronisation protocols

(16)

I

MPLEMENTATION

:

TESTS

● 3 links at 1.6 Gbit/s are operational

➥ several days of work without errors ➥ synchronisation of 2 links with a

time offset is demonstrated

● Large FPGA (ALTERA APEX400KE) is functional

➥ 500 I/O channels

➥ Ball Grid Array mounting technique ➥ operation at 40 MHz

(17)

C

ONCLUSIONS AND PROSPECTS

● The performance of the L0 Muon Trigger processor is very close to the

Références

Documents relatifs

Then, a physical algorithm is applied to give the trigger decision and a L0Block data is constructed.. The L0DU is built to be flexible : downscaling of L0 trigger condition,

From this model, we generate the relational database used to configure the software emulator of the Level-0 muon trigger and extract the implementation data required by

The principle of operation is based on the registration of x-rays induced by ions of accelerator beam interacting with a thin tungsten wire positioned on the beam path.. The

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des

Results: We have developed a computational approach to contextualise logical models of regulatory networks with biological measurements based on a probabilistic description

Unit´e de recherche INRIA Rocquencourt, Domaine de Voluceau, Rocquencourt, BP 105, 78153 LE CHESNAY Cedex Unit´e de recherche INRIA Sophia-Antipolis, 2004 route des Lucioles, BP

[r]

The business valuation model in this thesis adopts three concepts of financial modeling, Monte Carlo simulation (probability-generated cash flow), Capital Asset Pricing Model,