36x48 vertical poster template

1
Muon Reconstruction in CMS A. Everett, C. Liu,
N. Neumeister, A. Svyatkovskiy, H. Yoo Purdue
University
The Compact Muon Solenoid Detector
Reconstruction Performance

• CMS uses three types of gaseous particle
  detectors for muon identification
• Drift Tubes (DT) in the central barrel region
  (?lt1.2)
• 4 layers per superlayer
• 2-3 superlayers per station
• 4 stations
• Cathode Strip Chambers (CSC) in the endcaps
  (0.8lt?lt2.4)
• 1 wire plane and 1 cathode plane with strips per
  gap
• 6 gaps per chamber
• 4 stations
• Resistive Parallel Plate Chambers (RPC) in barrel
  and endcap
• 1-2 RPC per DT
• 1 RPC per CSC

Momentum resolution vs. psudorapidity and
momentum for Tracks from the silicon tracker,
Standalone Muons from the muon system, and Global
Muons from the silicon tracker and muon system.

• Resolution Measurement from Monte Carlo
• Calculate reconstruction momentum resolution as
  the relative difference of reconstructed track
  momentum to Monte Carlo generated track momentum
• At low energies, the Silicon Tracker dominates
  the momentum measurements
• At higher energies, the increased lever arm of
  the muon chambers causes the Standalone track
  momentum to dominate the momentum resolution
• Over all energies, the combination of Tracker and
  Standalone tracks (Global tracks) gives the best
  momentum resolution

Reconstruction Algorithms

• Muon Track Reconstruction
• Local Reconstruction of hits and segments in the
  muon chambers
• Reconstruction of the track in the muon system
  (Standalone Muon)
• Reconstruction of the track combining information
  from the tracker and the muon system (Global Muon)

Efficiency vs. Pseudorapidity for Standalone
Muons (left) and Algorithmic Efficiency for
Global Muons (right) for five momentum samples of
single muons

• Global Muon Reconstruction
• Global Muon Tracks combine information from
  whole tracking system of the CMS detector
• Use standalone muon parameters to select a region
  of interest in the silicon tracker
• Select tracks from silicon tracker
• Offline assume all silicon tracker tracks
  already reconstructed
• Online create a trajectory seed and grow a
  trajectory using standard Kalman algorithm
• Track Matching match tracker tracks to
  standalone tracks
• Global fit of track parameters using hits from
  tracker track and standalone muon track

• Standalone Muon Reconstruction
• Based on the Kalman Filter technique
• Seed state estimation
• Offline estimated from local segments
• Online estimated from L1 Trigger
• Inside-Out pattern recognition
• Middle point of segments avoids seed bias
• Outside-In track reconstruction
• Segments for pattern recognition and hits for
  Kalman Filter trajectory update
• Vertex Constraint extrapolate track to the point
  of closest approach to the beam line beam spot
  is constrained to be a point on the track

• Efficiency Measurement from Monte Carlo
• Calculate reconstruction efficiency as the ratio
  of reconstructed events to Monte Carlo generated
  events
• Standalone and Global Muon efficiencies near 100
  in entire detector
• The efficiency valleys are due to well understood
  boundaries between physical structures

Performance with Cosmic Muon Events

• Muon Identification
• Inside-Out strategy to complement offline
  Standalone and Global Muon Collections
• Look at all tracker tracks
• Check energy deposits in calorimeters and muon
  chambers
• Determine if deposits are consistent with a muon
  hypothesis

• Cosmic Muon Reconstruction
• Special reconstruction algorithms developed for
  cosmic ray muons
• Useful for detector alignment and calibration
• Useful to distinguish between muons from
  collisions, cosmic rays, beam-halo, etc.

Momentum resolution (Mean, left and Sigma, right)
vs. momentum for reconstructed muons from data
collected during CRAFT.

• Momentum Resolution Measurement from Cosmic Muon
  Data
• Calculate reconstructed muon momentum resolution
  using data collected from the Cosmic Run at Four
  Tesla (CRAFT) which ran with a magnetic field of
  3.8 T
• Compare the resolution from data with the
  resolution calculated using Cosmic Muon Monte
  Carlo

Illustration of differences between muons from
(a) collisions, (e) beam halo, and (b)(c)(d)(f)
cosmic rays