Summary of PRS/Muon
Activities
D.Acosta
University of Florida
Focus of Work
L2 and L3 higher level triggers, in preparation for the DAQ TDR
due out next year
è
Muon-Tracker matching
Staging scenarios
Assume initial luminosity of 2×1033
è DAQ input bandwidth reduced to 25 or 50 kHz
p Apply safety factor of 3, and split equally among muon,
e/gamma, and jet triggers (so BW could be as low as 2.5 kHz)
è No ME 4
è No RPC trigger in endcap
è No readout for inner part of ME 1/1
p Originally was strip “OR”
è
Production
Many headaches for INFN production centers because of Objectivity
problems and ORCA crashes
è Somewhat under control for ORCA5, samples now available and
results are coming out
è
EMU Meeting, October 2001
2
Darin Acosta
Plan for Large 12M Event Production
ECAL e/gamma group wants it for a calibration study
è Production will start ASAP at regional centers (including Florida)
è Production will not include pile-up (just single min bias events)
p Otherwise production requires too much effort
è Events are unbiased, so a good cross check of our rates and
punch-through
p From unbiased samples in Florida, I’ve seen that the trigger
rate is about the same whether you explicitly do the pile-up in
CARF or whether you just scale the single event rate.
è Sample size prohibits storing all events in database, so we only
get Ntuples
p So we get just one shot at this
è PRS group is finalizing the Ntuple code
è Smaller 0.5Mevt sample will be kept in a database for further
studies
è
EMU Meeting, October 2001
3
Darin Acosta
L1 single muon trigger rates
L=2·1033 cm -2 s-1
(ORCA 5.1.2) eff= 96.6%
eff= 96.3%
25-Sep-01
1.4 kHz
(pTcut= 20 GeV/c)
M.Fierro/HEPHY
7
L1 di-muon trigger rates, pT,2 ≥ 4 GeV/c
L=2·1033 cm -2 s-1
25-Sep-01
M.Fierro/HEPHY
8
Playing the game of thresholds
2.0 kHz
2.5 kHz
‘Iso-rate’ lines for
single muon and di-muon
Minimum Bias events
2.0 kHz
2.5 kHz
Efficiencies for
W→µ and Ζ/γ→µµ signals
for a fixed L1output bandwidth
25-Sep-01
M.Fierro/HEPHY
9
Combined Rate Results
L= 2·1033 cm-2 s-1
DAQ Bandwidth = 50kHz
(L1 µ = 6 kHz)
(muon, calo)
cuts
(GeV)
Muon
Ind.
Rates
(kHz)
16
2.5
Di-muon
(4, 4)
0.73
µ ⊕ Isol e/γ
(6, 8)
µ ⊕ τ jets
µ ⊕ Jets
Cum.
Rates
(kHz)
(muon, calo)
cuts
(GeV)
Ind.
Rates
(kHz)
18
1.9
0.65
(6, 6)
0.17
1.90
1.65
(8, 15)
(6, 65)
0.61
0.36
(6, 100)
0.18
0.01
L1 Total Rate
25-Sep-01
DAQ Bandwidth = 37.5 kHz
(L1 µ = 4 kHz)
5.2
Cum.
Rates
(kHz)
DAQ Bandwidth = 25 kHz
(L1 µ = 2.5 kHz)
(muon, calo)
cuts
(GeV)
Ind.
Rates
(kHz)
Cum.
Rates
(kHz)
18
1.9
0.14
(8, 8)
0.05
0.03
0.13
0.10
(14, 20)
0.02
0.01
(8, 70)
0.32
0.23
(14, 75)
0.10
0.04
(8, 100)
0.15
0.01
(14, 100)
0.06
0.00
2.4
2.0
M.Fierro/HEPHY
14
L3 Seed Generation
Start from L2 reconstructed muons:
•
Get muon trajectory at innermost muon station
– from local muon reconstruction
•
Propagate to outer tracker surface
– use GeaneWrapper
•
Rescale errors
– open window for track reconstruction
•
Find start layer(s) inside tracker
– 2 options : outside-in , inside-out
•
Get compatible hits from the layer
– if there are no compatible hits go to next layer
•
Create one or more seeds for each L2 muon
– set max. number of seed layers for each L2 muon (default = 3)
Norbert Neumeister
HEPHY Vienna
Muon PRS meeting
CERN, 11th September 2001
9
Efficiency
Global Efficiency
1
0.9
0.8
0.7
L1
L2
0.6
0.5
L3
0
Norbert Neumeister
HEPHY Vienna
0.5
1
1.5
Muon PRS meeting
CERN, 11th September 2001
2
2.5
µ
|η |
13
L2 pT Resolution
L2 : 1/pT resolution
barrel
overlap
Chi2 / ndf = 631.5 / 13
Constant = 4147
4500
Chi2 / ndf = 320.9 / 20
Constant = 1206
1400
Mean
= 0.004933
Sigma
= 0.1062
endcaps
Mean
= 0.004511
Chi2 / ndf = 1094 / 25
3500
Sigma = 0.1696
4000
3500
Constant = 3001
Mean
= 0.03827
Sigma = 0.2041
1200
3000
1000
2500
800
2000
600
1500
400
1000
200
500
3000
2500
2000
1500
1000
500
0
-1 -0.8 -0.6 -0.4 -0.2 -0
Norbert Neumeister
HEPHY Vienna
0.2 0.4 0.6 0.8
1
0
-1 -0.8 -0.6 -0.4 -0.2 -0
0.2 0.4 0.6 0.8
Muon PRS meeting
CERN, 11th September 2001
1
0
-1 -0.8 -0.6 -0.4 -0.2 -0
0.2 0.4 0.6 0.8
19
1
L3 pT Resolution
L3 : 1/pT resolution
barrel
overlap
Chi2 / ndf = 828.4 / 25
Constant = 3232
3500
Mean
= -0.0005702
Sigma
= 0.01452
endcaps
Chi2 / ndf = 221.4 / 27
Chi2 / ndf = 499.8 / 27
Constant = 693.5
800
700
Constant = 2659
Mean
= -0.0007918
Sigma
= 0.01855
3000
Mean
= -0.00469
Sigma
= 0.0235
2500
3000
600
2500
2000
500
2000
400
1500
1500
300
1000
1000
200
500
100
0
-0.1
500
-0.05
0
Norbert Neumeister
HEPHY Vienna
0.05
0.1
0
-0.1
-0.05
0
0.05
Muon PRS meeting
CERN, 11th September 2001
0.1
0
-0.1
-0.05
0
0.05
0.1
20
Muon: Lvl-3 status
n
With Lvl-3 resolution: rate given by ~ prompt muons
u
Moving on: how to reduce the rate further:
l Rate is mainly heavy flavors (b/c) (~ 100 Hz)
l Will look at isolation
l W/Z rate: 15 Hz at PT>20 GeV/c
P. Sphicas
PRS Report
CMS Annual Review
September 18, 2001
19
Summary
Group is very active, but not much U.S. involvement
beyond L1
Code and tools are in pretty good shape
è
First pass at reconstruction software
Don’t need to know C++ to participate, you can work
from standard Ntuples produced by the group
We need more help! Lots of staging studies to do,
HLT studies, etc.
Meetings are bi-weekly, you can connect from your PC
using VRVS software
è
Next one is this Tuesday, 10:30 EDT
EMU Meeting, October 2001
4
Darin Acosta