that location

bunch crossings (events)

pipeline

integrity

For data samples the available information is solely that coming from the reconstruction, but for simulation samples you can request both the reconstructed values of observables/variables and the generator-level (truth) values.

Selections made at these last stages are also defined by users. Their outputs are saved to disk and later used to produce data files that will contain only the information that interests users.

pipeline

whereby the selection of candidates made in the second stage of the high level trigger, HLT2, is saved to disk and used directly by analysts, with no further offline reconstruction by Brunel.

But, thanks to an enormous effort improving the reconstruction software both online and offline in between Run 1 and Run 2, the two reconstructions now perform identically.

If you don’t know the stripping line you need, it’s usually best to ask the stripping coordinators of the working group you’ll be presenting your work to.

Brunel application

the data flow and the associated applications look like this:

Gauss is responsible

f most important file formats used in LHCb

µDST format was designed to save space by storing only the information concerning the build candidates

data flow designed to maximise the data-taking efficiency and data quality.

the data must be processed very quickly because it’s arriving very quickly, and the data is complex so there’s a lot that can be done with it.

filter the data and try to keep only the events which contain something interesting

optimisation

we can assign a momentum estimate to a track based on its curvature,

You can find the source code of these lessons in the lhcb/first-analysis-steps repository,

Prerequisites

survive from

Charmless backgrounds

Potential fit bias is checked byperforming an ensemble test comprising 1000 pseudo-experiments, where signal is taken from the correspond-ing MC sample, and the PDF shapes are used to gen-erate background events. We obtain a Gaussian nor-malized residual distribution of unit width, and add itsmean and uncertainty in width in quadrature to cal-culate the systematic error.

The uncertainties due to PDF shapesare estimated by varying all the fixed shape parame-ters by their errors.

control sample

as obtained fromthe charged decay.

PDF shape parameters

we multiply the denominator by a factor of 12to account for K0 → K0S

oss in efficiency on top of the smallsignal yield

since we would need to tag the recoiling B candidate forthat, causing further loss in efficiency on top of the smallsignal yield.

eventcategory

event category

The signal yield is obtained with an unbinned extendedmaximum-likelihood fit to the three variables Mbc, ∆E,and O′NN.

We apply a loose criterionONN > −0.5 to reduce the continuum background

B thrust axis

observables based on event topology

misreconstructed

χ2 value

B-vertex fit

In the case of multiple B candidates, wechoose the candidate with the lowest χ2 value for theaforementioned B-vertex fit.

The Mbcrequirement corresponds to approximately (−16σ, +3σ)in resolution around the nominal B mass [12], and the∆E requirement denotes a ±10σ window around zero

further study

a maximum-likelihood fi

We perform a fit for each B candidate, constrainingits decay products to originate from a common vertex

the beam-energy-constrained mass

distance of closest approach

candidates