Aspects of the LVL2 Analysis

1
Aspects of the LVL2 Analysis
Alberica Toia Ingo Frölich, Adrian Gabriel,
Daniel Kirschner, Wolfgang Kühn, Jörg Lehnert,
Erik Lins, Markus Petri, Jim Ritman, Daniel
Schäfer, Michael Traxler II. Physikalisches
Institut University of Giessen

• Functionality(do the boards behave as they
  should ?)
• RICH IPU
• TOF IPU
• Shower IPU
• Matching Unit
• Performance(do the boards behave reasonably ?)
• Efficiency
• Reduction Factor

2
RICH IPU
ring search mask

• Pattern reconstruction (96x96 pad plane)
• Ring recognition
• fixed 80 pads mask (ring/veto region)
• local maximum search

uncorrelated emulation rings
Data Nov01C-C 1.5AGeV full field

• Discrepancy 0.4

3
TOF IPU
time difference

• On-line calibration of data
• Selection of particles in a time window

IPU-offline f correlation
f IPU ()
t (ns)
IPU-offline time correlation
without start correction
f difference
t IPU (ns)
Df 0.22 s 0.66
with start correction
f ANA ()
Df ()
Data Jan02cosmics

• Good time and angular resolution
• Discrepancy 0.2

t ANA (ns)
Data Aug01C-C 1.5AGeV full field
4
Shower IPU

• Pattern reconstruction (3 32x32 pad planes)
• Shower recognition
• Increasing of charge on consecutive chambers
• Local maximum search

Data Mar02cosmics

• Only local maximum
• Resolution 1 pad
• Discrepancy due to comunication between the 2
  boards

5
Matching Unit

• Combines position information of inner and outer
  IPUs to select lepton candidates (q, f, p)

Data Nov01C-C 1.5AGeV full field
Nent 6420
Nent 9191
Nent 6453
6
Matching Unit II
Data Nov01C-C 1.5AGeV full field
Momentum
Matching Unit

• Calculates the momentum with a look-up table
  p(q, Dq, f)
• Combines leptons into un-like sign pairs with a
  given opening angle and invariant mass range

Matching Unit Emulation
Invariant Mass
p (MeV/c)
EMU
Opening Angle
EMU
m (MeV/c2)
Opening angle ()
7
Performance
Trigger condition evts evt gt 1
ring 10.5 evt gt 1 lepton 2.9 evt gt 2
rings 4.6 evt gt 2 leptons 1.7 evt gt 1
dilepton 0.3
-gt RICH IPU is the most selective element of
LVL2 -gt further characterization in terms of
RICH IPU
!!! NO Shower IPU electrons gtgt positrons !!!
Data Nov01C-C 1.5AGeV full field
1) EfficiencyHow many leptons are recognized as
rings ?(definition of leptons) 2) Reduction
FactorWhere are the fakes ?Can they be reduced ?
8
Efficiency with simulations
CC _at_1.9 AGeV e-
Photon distribution
Photon distribution efficiency loss at low
photon multiplicity (low polar angle)
leptons
87.6 rings
found by the IPU
photons per electron
p (MeV/c)
q (degrees)
9
Efficiency with beam
p (MeV)

• Reference System offline analysis leptons
• LVL2 conditions
• 1 hit in RICH
• 1 hit in META
• Df lt 15

q (degrees)
Data Apr01C-C 1.5AGeV full field
77057 leptons found by the LVL2
3618 leptons found by the offline analysis
RICH IPU LVL2 integrated
eff 85.3 84.7 event eff 87.3 86.6 fakes/e
vt 0.24 0.19
Fake candidates ???
leptons
10
Reduction factor
How to reduce fakes?
Pad distribution
simulations
data
rings found by the IPU
rings found by the IPU
59 corresponding to RICH rings
37 corresponding to RICH rings
12 corresponding to leptons
3 corresponding to leptons
pads per ring
pads per ring
Data Apr01C-C 1.5AGeV full field

• Fakes concentrated at low pad multiplicity
• Increase of thresholds would cost an efficiency
  loss

11
Summary and outlook

• Consistency
• RICH IPU
• good agreement hardware-emulation
• TOF IPU
• hardware-software agreement compatible with
  detector performance
• Shower IPU
• good agreement hardware-emulation
• part of the algorithm still to be implemented
• Matching Unit
• good agreement hardware-emulation
• part of the algorithm still to be implemented
• Performances
• Reduction factor enables a high event
  suppression
• Single lepton efficiency
• simulations bias on digitizer and event
  generator
• data which reference system ???
• -gt necessity of a calibration reaction (pp -gt
  pph)

12
MU invariant mass
un-like sign pairs
like sign pairs
un-like sign pairs event mixing
m(MeV/c2)
13
MU invariant mass
All pairs
Like-sign pairs
unlike-sign pairs Event mixing
unlike-sign pairs