Progress update on muon combinatorial background [ - - PowerPoint PPT Presentation

progress update on muon combinatorial background
SMART_READER_LITE
LIVE PREVIEW

Progress update on muon combinatorial background [ - - PowerPoint PPT Presentation

Progress update on muon combinatorial background [ oliver.lantwin@cern.ch ] 11th SHiP Collaboration Meeting, cern 8th June 2017 Oliver Lantwin , Kostas Petridis Introduction Update of combinatorial background studies with Oliver Lantwin (icl)


slide-1
SLIDE 1

Progress update on muon combinatorial background

Oliver Lantwin , Kostas Petridis

[oliver.lantwin@cern.ch]

11th SHiP Collaboration Meeting, cern 8th June 2017

slide-2
SLIDE 2

Introduction

Update of combinatorial background studies with

◮ New detector geometry and shield ◮ Updated kinematics for toy generator

Oliver Lantwin (icl) 11th SHiP Collaboration Meeting, cern Introduction 2/19

slide-3
SLIDE 3

Introduction Muons from Muon Background Study of muon kinematics at the tracking stations Toy-mc Results Conclusion

slide-4
SLIDE 4

Simulation procedure

NB: Closely following Kostas’s procedure used for tp with some recent departures (see later slides)

◮ Use muons with well-reconstructed tracks from muon background ◮ Form all possible combinations ◮ Use distributions of position, momenta to seed a fast simulation

◮ Use data-driven methods to estimate kinematics ◮ Generate according to these kinematics Oliver Lantwin (icl) 11th SHiP Collaboration Meeting, cern Introduction 4/19

slide-5
SLIDE 5

Muons pairs used as seed

◮ Updated to 10× the full muon sample (9× with randomised φ)

◮ 29 muons → 406 combinations

◮ Current geometry ◮ Current reconstruction

Oliver Lantwin (icl) 11th SHiP Collaboration Meeting, cern Muons from Muon Background 5/19

slide-6
SLIDE 6

Hit distribution of muons (used as seed)

X 1st state (cm)

  • 500
  • 400
  • 300
  • 200
  • 100

100 200 300 400 500 Y 1st state (cm)

  • 500
  • 400
  • 300
  • 200
  • 100

100 200 300 400 500

CombMuXYstate Entries 29 Mean x

  • 24.99

Mean y 62.98 Std Dev x 115 Std Dev y 170.9 Integral 1.275e+05 Skewness x 0.04238 Skewness y 0.03767 127452

1000 2000 3000 4000 5000 6000 7000 8000 9000

CombMuXYstate Entries 29 Mean x

  • 24.99

Mean y 62.98 Std Dev x 115 Std Dev y 170.9 Integral 1.275e+05 Skewness x 0.04238 Skewness y 0.03767 127452

XY Oliver Lantwin (icl) 11th SHiP Collaboration Meeting, cern Muons from Muon Background 6/19

slide-7
SLIDE 7

pt–p distribution of muons (used as seed)

pT [GeV] 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 p[GeV] 50 100 150 200 250 300

CombMuPtPstate Entries 29 Mean x 0.2509 Mean y 10.01 Std Dev x 0.2518 Std Dev y 14.78 Integral 1.275e+05 Skewness x 1.382 Skewness y 2.701 127452 χ2/nd f 4.602e+08 / 997 p1 −4.068 ± 0.0119 p2 −0.09252 ± 0.0002628 p3 11.24 ± 0.005143

10000 20000 30000 40000 50000 60000 70000

CombMuPtPstate Entries 29 Mean x 0.2509 Mean y 10.01 Std Dev x 0.2518 Std Dev y 14.78 Integral 1.275e+05 Skewness x 1.382 Skewness y 2.701 127452 χ2/nd f 4.602e+08 / 997 p1 −4.068 ± 0.0119 p2 −0.09252 ± 0.0002628 p3 11.24 ± 0.005143

pT p ◮ Simple 2D Exponential fjt (no cross-terms, assume p, pt uncorrelated) ◮ Fit function chosen for tp because of lack of statistics → investigate

using something more realistic now?

Oliver Lantwin (icl) 11th SHiP Collaboration Meeting, cern Muons from Muon Background 7/19

slide-8
SLIDE 8

The toy-mc

◮ Sample muon positions from histogram of seed muons ◮ Sample muon momenta from fjt ◮ Randomise φ of muon direction → Uniform distribution? Assumption

justifjed? → Study kinematics more closely!

Oliver Lantwin (icl) 11th SHiP Collaboration Meeting, cern Muons from Muon Background 8/19

slide-9
SLIDE 9

Setup

◮ Generate 10 000 000 muons with particle gun, uniform in 3 < η < 8 and

2 GeV < p < 400 GeV and −π < φ < π

◮ Hans’s shield confjguration ◮ Study φ as a function of p, pT for muons hitting the tracker

P[GeV] 50 100 150 200 250 300 350 400 PT [GeV] 5 10 15 20 25 30 35 40

htemp Entries 1e+07 Mean x 201 Mean y 3.972 Std Dev x 114.9 Std Dev y 6.1 Integral 1e+07 Skewness x
  • 0.000686
Skewness y 2.424 1e+07

1 10 102 103 104 105

htemp Entries 1e+07 Mean x 201 Mean y 3.972 Std Dev x 114.9 Std Dev y 6.1 Integral 1e+07 Skewness x
  • 0.000686
Skewness y 2.424 1e+07

µ±

Figure: Generated momentum distribution

◮ NB: signifjcantly higher momenta and transverse momenta than in actual

muon spectrum, beyond kinematic limit

Oliver Lantwin (icl) 11th SHiP Collaboration Meeting, cern Study of muon kinematics at the tracking stations 9/19

slide-10
SLIDE 10

p-pT distribution before and after shield

P[GeV] 50 100 150 200 250 300 350 400 PT [GeV] 2 4 6 8 10 12 14 16 18 20 mu_ppt_original

Entries 807288 Mean x 286.1 Mean y 5.728 Std Dev x 88.39 Std Dev y 2.886 Integral 8.073e+05 Skewness x
  • 0.7555
Skewness y 1.034 33 807255

100 200 300 400 500 600

mu_ppt_original

Entries 807288 Mean x 286.1 Mean y 5.728 Std Dev x 88.39 Std Dev y 2.886 Integral 8.073e+05 Skewness x
  • 0.7555
Skewness y 1.034 33 807255

µ±

P[GeV] 50 100 150 200 250 300 350 PT [GeV] 2 4 6 8 10 12 14 16 18 20

p_pt

Entries 403850 Mean x 200.4 Mean y 4.193 Std Dev x 78.58 Std Dev y 2.577 Integral 4.038e+05 Skewness x
  • 0.4358
Skewness y 0.9441 403850

50 100 150 200 250 300 350 400 450

p_pt

Entries 403850 Mean x 200.4 Mean y 4.193 Std Dev x 78.58 Std Dev y 2.577 Integral 4.038e+05 Skewness x
  • 0.4358
Skewness y 0.9441 403850

µ−

Figure: Momentum and transverse momentum of muons creating hits in the tracking station: left: at generation; right: at the tracking station

Oliver Lantwin (icl) 11th SHiP Collaboration Meeting, cern Study of muon kinematics at the tracking stations 10/19

slide-11
SLIDE 11

φ distribution

φ

  • 3
  • 2
  • 1

1 2 3 2000 3000 4000 5000 6000 7000

phi

Entries 403850 Mean

  • 0.0212

Std Dev 1.983 Underfmow Overfmow Integral 4.038e+05 Skewness 0.01347

µ−φ ◮ Results for muons, for anti-muons results are virtually the same with φ

shifted by π

◮ φ distribution clearly not uniform

Oliver Lantwin (icl) 11th SHiP Collaboration Meeting, cern Study of muon kinematics at the tracking stations 11/19

slide-12
SLIDE 12

φ-pT and φ-P distributions

φ

  • 3
  • 2
  • 1

1 2 3 P[GeV] 50 100 150 200 250 300 350

phi_p

Entries 403850 Mean x
  • 0.0212
Mean y 200.4 Std Dev x 1.983 Std Dev y 78.58 Integral 4.038e+05 Skewness x 0.01347 Skewness y
  • 0.4358
403850

20 40 60 80 100 120 140 160 180

phi_p

Entries 403850 Mean x
  • 0.0212
Mean y 200.4 Std Dev x 1.983 Std Dev y 78.58 Integral 4.038e+05 Skewness x 0.01347 Skewness y
  • 0.4358
403850

µ−φ − P

φ

  • 3
  • 2
  • 1

1 2 3 PT [GeV] 2 4 6 8 10 12 14 16 18 20

phi_pt

Entries 403850 Mean x
  • 0.0212
Mean y 4.193 Std Dev x 1.983 Std Dev y 2.577 Integral 4.038e+05 Skewness x 0.01347 Skewness y 0.9441 403850

50 100 150 200 250 300 350 400 450

phi_pt

Entries 403850 Mean x
  • 0.0212
Mean y 4.193 Std Dev x 1.983 Std Dev y 2.577 Integral 4.038e+05 Skewness x 0.01347 Skewness y 0.9441 403850

µ−φ − P

◮ Some clear hotspots visible.

Oliver Lantwin (icl) 11th SHiP Collaboration Meeting, cern Study of muon kinematics at the tracking stations 12/19

slide-13
SLIDE 13

The toy-mc

◮ Sample muon positions from histogram of seed muons ◮ Sample muon momenta from fjt ◮ Sample φ from p-pt-φ histogram in slice corresponding to generated p-pt

→ further studies into implicit assumptions ongoing

◮ Generate 1958* muons → 1915903 pairs

*2 events are discarded because their pt > p Oliver Lantwin (icl) 11th SHiP Collaboration Meeting, cern Toy-mc 13/19

slide-14
SLIDE 14

Generated pt–p distribution of muons (from toy-mc)

pT [GeV] 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 p[GeV] 50 100 150 200 250 300

CombGenMuPtPstate Entries 1953 Mean x 0.282 Mean y 11.49 Std Dev x 0.2354 Std Dev y 11.37 Integral 1953 Skewness x 1.829 Skewness y 2.261 1953

50 100 150 200 250 300 350 400

CombGenMuPtPstate Entries 1953 Mean x 0.282 Mean y 11.49 Std Dev x 0.2354 Std Dev y 11.37 Integral 1953 Skewness x 1.829 Skewness y 2.261 1953

pT p

◮ Note: Sample muon momenta from fjt (i.e. interpolate histogram) up to

100 GeV

Oliver Lantwin (icl) 11th SHiP Collaboration Meeting, cern Results 14/19

slide-15
SLIDE 15

Generated φ distribution of muons (from toy-mc)

φ

  • 3
  • 2
  • 1

1 2 3 10 20 30 40 50 CombGenMuPhi

Entries 1953 Mean 0.06746 Std Dev 1.772 Underfmow Overfmow Integral 1953 Skewness

  • 0.05129

µ−φ

Oliver Lantwin (icl) 11th SHiP Collaboration Meeting, cern Results 15/19

slide-16
SLIDE 16

Comparison of seed muons vs. toy-mc muons

IP to target (cm) 200 400 600 800 1000 1200 1400 1600 1800 2000 Entries 0.0005 0.001 0.0015 0.002 0.0025 CombGenMuTargIP

Entries 1906128 Mean 377.6 Std Dev 412.8 Underfmow Overfmow 2.177e+04 Integral 1.884e+06 Skewness 1.42

IPtar

Oliver Lantwin (icl) 11th SHiP Collaboration Meeting, cern Results 16/19

slide-17
SLIDE 17

Comparison of seed muons vs. toy-mc muons

doca(µ, µ)(cm) 20 40 60 80 100 120 140 160 180 200 Entries 0.002 0.004 0.006 0.008 0.01 CombGenMuDoca

Entries 1906128 Mean 161.4 Std Dev 122.2 Underfmow Overfmow 6.411e+05 Integral 1.265e+06 Skewness

  • 0.3702

Doca

Oliver Lantwin (icl) 11th SHiP Collaboration Meeting, cern Results 17/19

slide-18
SLIDE 18

Comparison of seed muons vs. toy-mc muons

vtxz(µ, µ)(cm)

  • 10000
  • 8000
  • 6000
  • 4000
  • 2000

2000 4000 6000 8000 10000 Entries 0.02 0.04 0.06 0.08 0.1 0.12 0.14 ×10−3 CombGenMuVtxZ

Entries 1906128 Mean 2355 Std Dev 2.611e+04 Underfmow 1.036e+05 Overfmow 1.753e+05 Integral 1.627e+06 Skewness

  • 0.005955

VtxZ

Oliver Lantwin (icl) 11th SHiP Collaboration Meeting, cern Results 18/19

slide-19
SLIDE 19

Conclusion

Update of combinatorial background studies with

◮ New detector geometry and shield ◮ Updated kinematics for toy generator

TODO

◮ Study relation between muon kinematics at generator and tracker to

prove that method used to sample φ OK

◮ Cross-check by combining muons from different hnl decays ◮ Comparison with hnl signal ◮ Scale results and look at effjciency of cuts

Estimation of computing time:

◮ Bottleneck: muon generation for seeding the toy-mc → a few days

◮ Done, unless we want even more statistics, but randomising φ only gets us so

far

◮ Toy-mc: O(minutes)

Oliver Lantwin (icl) 11th SHiP Collaboration Meeting, cern Conclusion 19/19