SPD software Status

1
SPD software Status

• By
• Bjørn S. Nilsen
• The Ohio State University
• And
• Tiziano Virgili

2
Outline

• Simulation code status
• Merged model
• Diffusion/Cluster analysis status
• Vertexing status
• New ITS Geometry

3
Merged Model

• Comparisons with Bari/Serno model show equivalent
  results.
• Different ionization energy used
• New Model uses 1micron step sizes
• Speed depends on step size (optimization to be
  done)
• Steps more compatible with diffusion
• Bari/Serno model computes pixel boundaries
• Present implementation is slow and could be a lot
  faster.

4
Diffusion/Cluster analysis status

• Nothing new since October.
• New student still learning C and ROOT/AliRoot
• Still need a proper comparison with Dubna model
  and new model.
• Still need to tune parameters to match test beam
  data.

5
Multiplicity in p-p

• Multiplicity reconstruction with the Silicon
  Pixel Detector

New problem in p-p respect to Pb-Pb the vertex
position is not always available. Also, in some
cases the reconstruction is not good. This
problem doesnt affect the cluster multiplicity.
8200 events p-p from the last PPR production
were analyzed. In the following the main results
are recalled.
6
Multiplicity in p-p

• Multiplicity in the SPD acceptance
• (hlt2 for layer 1 hlt1.5 for layer 2 and
  tracklets )

7
Multiplicity in p-p

• Multiplicity in the central unit of h

8
Multiplicity in p-p

• Efficiency
• (Only events with reconstructed vertex are used.)

9
Multiplicity in p-p

• Multiplicity distribution (points correspond to
  the tracklet multiplicity.)

10
Multiplicity in p-p

• Vertex resolution as a function of multiplicity.
  The reconstruction is missing for most of the
  events at very low multipli-city. Also, there are
  some events with wrong reconstruction!

11
Multiplicity in p-p

• Correlation between the tracklet and cluster
  multiplicity

Correlation between the tracklet mult. evaluated
with true and reconstructed vertex position.
12
Multiplicity in ALICE
C The events with wrong vertex determination
can be identified from the correlation beteween
cluster and tracklet multiplicity!
13
New ITS Geometry

• SPD test-beam geometry introduced.
• Coding just started last week.
• This is going to be a BIG JOB.
• A fully parameterized geometry is being developed.

14
CONCLUSIONS

• Progrees with changes to SPD model slow.
• Vertexing, all the methods give good results.
• The tracklets give a better measurement of
  multiplicity, provided the vertex is well
  recon-structed.
• Events with wrong vertex determination can be
  identified from the correlation between cluster
  and tracklet multiplicity.
• New ITS Geometry started.