1. Flavour Tagging performance in LHCb Marc Grabalosa Gándara (14/12/08)
DISCRETE '08 Symposium on Prospects in the Physics of Discrete Symmetries 
11–16 December 2008, IFIC, Valencia, Spain
Flavour Tagging
performance in LHCb
Marc Grabalosa Gándara (14/12/08)
on behalf of the LHCb collaboration

2. Motivation Unitarity Triangles
Bd0  p+ p-
Bd0  r p
Bd0  DK*0
BS0  DSK
Bd0  D* p, 3p
BS0  DS p
Bd0  J/y KS0
BS0  J/y f
LHCb is a 2nd generation precision experiment coming after B-Factories and Tevatron
Improve precision on g and other CKM parameters
Many measurments require the knowledge of the initial flavour of the B meson

3. Importance of tagging 3
Bs oscillation frequency, phase and ΔΓs ( BsDs, J/ΨΦ, J/Ψη, ηcΦ )
Measure the CKM parameters
 from Bd0–+
 with BdJ/KS as a proof of principle (  from bs penguin )
 in various channels, with different sensitivity to new physics:
Time-dependent CP asymmetry of BsDs-K+ and Ds+K-
Time dependent CP asymmetries of Bd π+ π- and Bs  K+K-
Comparison of decay rates in the BdD0(K+π-,K-π+,K+K-)K*0 system
Comparison of decay rates in the B-D0(K+π-,K+π-π+π-)K- system
Dalitz analysis of B-  D0(KSπ-π+)K- and Bd  D0(KSπ-π+)K*0
Rare B decays
Radiative penguin Bd  K* , Bs  Φ, Bd  
Electroweak penguin Bd  K*0 +-
Gluonic penguin Bs  ΦΦ, Bd  ΦKs
Rare box diagram Bs +-
Bc , b-baryon physics + unexpected !
TAGGING REQUIRED

4. LHCb Overview Requirements for CP measurements in B the sector are
Good particle Id,
excellent tracking and vertexing
B flight path of the order 5-10mm
Tracking:
δp/p= 0.35% to 0.55%
Vertexing:
~10μm transverse plane and ~60μm in z
Expected Impact parameter resolution
σIP =14μm + 35μm/pT
Calorimeter resolution:
σE/E = 1% + 10%√E (E in GeV)‏
RICH:
Particle identification, important to distinguish between Kaon and pions.

5. Flavour Tagging 5
Taggers:
muons
electrons
kaons
vertex charge
kaons or pions (when Bd,u)‏
Same Side
Opposite Side
Signal B
Opposite B p
K+ ()‏
l - K-
Qvrtx OS SS
If several candidates for the same tagger exist  Select the one with highest Pt.
Tagging efficiency
Wrong tag fraction
Effective efficiency
Taggers make individual decisions about the flavour with varying accuracy, which is evaluated by a NNet.

6. Opposite-side tagger (OS)6
Tagging objects from b → c → s chain.
Kinematic and geometrical variables (IPS, P, Pt,...) show a dependence in purity of right vs wrong tags  CUTS
pT of m’s:
IP of K’s:

7. OS Vertex Charge tagger7
Use long tracks to build a 2-seed vertex after some kinematic cuts
Use a NN to select good candidate (2-seed) to SV
Other tracks are added iteratively
Weighted charge can be used as a tagger
All vtx
Good vtx
seed SV
All trakcs
Tracks coming from b
Typical performance: e = 43% w = 42% eeff = 1.14%

8. Same-side tagger (SS) Particle selection cuts Proximity to signal b8
Particle selection cuts
Proximity to signal b
Hadron from fragmentation (K )
or B** decay ()
 from B*
 from fragmentation
Other sources
Wrong tagging
K from fragmentation
Other sources
Wrong tagging
Typical performance: e = 25.5% w = 35.6%
eeff = 2.13%
Bs → Ds K
B0 → + -

9. Taggers 9
The tag (b or bbar) is decided by the charge of the tagging object
Combine the taggers to obtain a final decision of the tag
Sort in 5 categories depending on the probability of the tag to be correct
Each method will give a tag and a category
(related with the reliability)
Combine Particle IDentification (PID)
Sort events based on the PID of the track ordering them in 
NN independent. Simple method
Has a lower efficency
Neural Net
Obtain a wrong tag fraction () for each event from the NN output
Has a higher efficency

10. Neural Net method 10
For each event, each tagger will give an  as a function of the NN output.
The wrong tag fraction is fit linearly on the Neural Net output.
Bs  J/ 
right
wrong
Opposite kaon
tagger(K) (NNet) = a0 + a1 NNet

11. Combination of taggers11
Each tagger will have its own  tagger (NNet).
The final probability for the event will be a combination of the tagger wrong tag fractions:
P+1 = (1-k)  e …
P-1 =  k (1-  e) …
If P(B) < 0.55 events is left untagged
Cat 1
Cat 2
Cat 3
Cat 4
Cat 5
To calculate the final combined effective efficiency, we bin the events in 5 categories (and treat them separately in the CP fits).
P(B)

12. PID based combination of taggers12
Form possible combinations according:
Sort all possible combinations, including the case when abs(sum)>1, according to the  estimated on a control channel (62 possible combinations, but only 41 non empty)‏
Bin events in 5 categories
Particle Identification (PID) Muons, electrons , kaons, kaons or pions SS , vertex charge
Sum of the individual tagger decisions (sum of charges)‏ abs(sum) > 1
Untagged
= 43.0%
Smaller 
= 32.3% 
Cat 5
Cat 4
Cat 3
Cat 2
Cat 1
PID combination

13. Results, ex. Bs  J/  Performance of taggers:13
Performance of taggers:
Combine all taggers to obtain the global effective efficency, which is the direct sum of εeff in the 5 tagging categories.
NNet εeff increases by ~20%

14. Performances for a few channels
eeff %
e %
w %
Bs→ Dsp
8.85 ± 0.18
60.7
30.9
Bd→ J/ψ K*
4.29 ± 0.09
53.2
35.8
Bd→ pp
5.52 ± 0.16
56.8
34.4
Bu→ J/ψ K+
4.11 ± 0.11
53.1
36.1
Differences can be due to different signal B spectra, trigger…

15. Control channels 15
Accumulate high statistics in various flavour-specific modes
 can be extracted by:
B±: just comparing tagging with observed flavour
Bd and Bs need fit of oscillation
Taggers can be calibrated using these control channels.
Channel
Yield/2 fb-1
B/S
dw / w ( 2fb-1 ) estimate
B+J/()K+
1.7 M
0.4
0.15%
B+D0p+
0.7 M
0.8
0.25%
B0J/()K*0(K+-)
0.2
0.2%
Bs Ds+ p-
0.08 M
0.3
0.7%
Bd0 D* - m+ n
9 M
0.05%
B+ D0 (*) m + n
3.5 M
0.6
0.1%
Bs Ds(*) m + n
2 M
0.1
0. 5%
Topology close to that of signal channels
Semileptonics:
High statistics
More difficult topology

16. Use of control channels16
B+ J/ K+ is a flavour specific channel
No true MC information needed
The  obtained in a given tagger for B+ J/ K+ can be used the same taggers in other channels
Control channels will allow to measure  directly from data, with the statistical accuracy required for physics measurements
B+  J/ K+
Bd  J/ Ks
right
wrong
right
wrong
Opposite kaon
tagger(K) (NNet)=a0 + a1 NNet

17. Mistag extraction for B0 → J/ Ks
One of the first measurements requiring flavour tagging of the B will be
sin 2 from B0JKS as a benchmark to demonstrate LHCb
capability in CP-asymmetry measurements
For the evaluation of the mistag rate, the following strategy, using B+JK+ and B0JK*0 as control channels, is foreseen:
With B+JK+ events determine for each tagger the dependence of the mistag rate on the kinematical properties of the tagger.
Combine these probabilities into a single probability per event.
Use this function to subdivide B0JK*0 and B0JKS events into samples of decreasing mistag-rate (tag categories).
Fit to flavour oscillations of B0JK*0 events, as a function of propertime, in each of the 5 samples, to measure the mistag rate per category. Use these 5 mistag rates in the CP fit of B0JKS events, also subdivided into 5 categories.

18. Fit to flavour oscillations of B0  J/ K*0 in 5 categories
Only signal events
considered here
Cat 3
Cat 2
Cat 1

19. Control channel check
from MC truth
from propertime fit
Results from propertime fit are compatible to MC truth.
In one year, 2/fb, with 215k events, s(sin2b)~0.02

20. Background on control channels20
Control channels will be used with data events, where full account of background has to be taken.
We have devised the strategies to cope with it.
For Bd+ J/ K*

21. Conclusions 21
Flavour tagging is a fundamental ingredient for B physics measurements in LHCb.
Control channels will allow to measure  directly from data, with the required statistical accuracy, taking into account many possible effects (backgrounds, trigger, etc.)
Expected effective tagging efficiency at LHCb is
~ 6 – 9 % for Bs and ~ 4 – 5 % for Bd,u channels