title	author	institute	date	theme	fonttheme	classoption
Measurement of \RDX in semileptonic $B$ decays and upgrade of the LHCb Upstream Tracker	Yipeng Sun	University of Maryland	Mar 1st, 2023	UMDPepsi	serif	aspectratio=169,dvipsnames

Outline

::: columns ::: {.column width=50%}

Preliminary measurement of \RDX
- Introduction
- Event selection
- Trigger emulation for MC
- Data/MC corrections
- Fit
- Systematics (WIP)

::: ::: {.column width=50%}

Upgrade of the LHCb detector
- Overview of the LHCb upgrade
- The Upstream Tracker

::: :::

Introduction

::: columns ::: {.column width=50%}

Preliminary measurement of \RDX
- Introduction \color{gray}
- Event selection
- Trigger emulation for MC
- Data/MC corrections
- Fit
- Systematics (WIP)

::: ::: {.column width=50%}

Upgrade of the LHCb detector
- \color{gray}Overview of the LHCb upgrade
- The Upstream Tracker

::: :::

The standard model and beyond

::: columns ::: {.column width=65%}

The standard model (SM) is a hugely successful theory
- A QFT describing interactions between fermions & bosons
- Allow arbitrary identical copies of leptons (except for interactions w/ Higgs) $\rightarrow$ lepton flavor universality (LFU)
- 3 generations of leptons: $e, \mu, \tau$

::: ::: {.column width=35%} \centering { width=80% } ::: :::

\vspace{1em} ::: columns ::: {.column width=70%} \begin{itemize}

\tightlist \item<2-> SM doesn't explain every experimental observation

\begin{itemize} \tightlist \item Matter-anti-matter asymmetry of the universe \item Evidence of dark matter from cosmological observations \item Demand new physics (NP) beyond the SM \end{itemize} \end{itemize}

::: ::: {.column width=30%}

\visible<2->{ \vspace{0.5em} \includegraphics[width=0.85\textwidth]{./slides-figures/darkmatter.png} }

::: :::

Precision measurements & tests of LFU

\vspace{-0.5em} \small\tightmargin

Testing SM with precision measurement
- Measure observables precisely
- Compare with precise theoretical predictions
- Deviations from SM predictions $\rightarrow$ hints to NP

\visible<2->{ \vspace{-1.5\baselineskip} \begin{itemize} \tightlist \item LFU has been tested to high precision, \textbf{no definite violation so far} \end{itemize} \vspace{-\baselineskip} }

\setlength{\leftmargini}{12pt} \vspace{-0.5em} ::: columns ::: {.column width=50%}

\visible<2->{ \begin{block}{LFU tests with $e, \mu$} \footnotesize \begin{itemize} \tightlist \item To \textbf{0.28%}: $Z$ decays \begin{itemize} \tightlist \item $\frac{\Gamma_{Z \rightarrow \mu\mu}}{\Gamma_{Z \rightarrow ee}} = 1.0009 \pm 0.0028$ \item LEP, \href{https://arxiv.org/abs/hep-ex/0509008}{\scriptsize\texttt{Phys. Rept. 427 (2006) 257}} \end{itemize}

\item To \textbf{0.8%}: $W$ decays \begin{itemize} \tightlist \item $\frac{\mathcal{B}(W \rightarrow e \neu_e)}{\mathcal{B}(W \rightarrow \mu \neu_\mu)} = 1.004 \pm 0.008$ \item CDF + LHC, \href{https://arxiv.org/abs/1809.06229}{\scriptsize\texttt{JPG: NPP, 46, 2 (2019)}} \end{itemize}

\item About \textbf{9.5%} (low-\qSq): $R_{K^{(*)}}$ \begin{itemize} \tightlist \item $\frac{\Gamma_{B \rightarrow K \mu\mu}}{\Gamma_{B \rightarrow K ee}} = 0.994^{+0.094}_{-0.087}$ (low-\qSq) \item LHCb, \href{https://indico.cern.ch/event/1187945/}{\scriptsize LHCb seminar (2022)} \end{itemize}

\end{itemize} \end{block} }

::: ::: {.column width=50%}

\visible<2->{ \begin{block}{LFU tests with $\mu, \tau$} \footnotesize \begin{itemize} \tightlist \item To \textbf{0.32%}: $Z$ decays \begin{itemize} \tightlist \item $\frac{\Gamma_{Z \rightarrow \tau\tau}}{\Gamma_{Z \rightarrow \mu\mu}} = 1.0019 \pm 0.0032$ \item LEP, \href{https://arxiv.org/abs/hep-ex/0509008}{\scriptsize\texttt{Phys. Rept. 427 (2006) 257}} \end{itemize}

\item To \textbf{1.3%}: $W$ decays \begin{itemize} \tightlist \item $\frac{\Gamma_{W \rightarrow \tau \neu_\tau}}{\Gamma_{W \rightarrow \mu \neu_\mu}} = 0.992 \pm 0.013$ \item ATLAS, \href{https://arxiv.org/abs/2007.14040}{\scriptsize\texttt{arXiv:2007.14040}} \end{itemize}

\item To \textbf{6.1%}: $D_s$ decays \begin{itemize} \tightlist \item $\frac{\Gamma_{D_s \rightarrow \tau\nu_\tau}}{\Gamma_{D_s \rightarrow \mu\nu_\mu}} = 9.95 \pm 0.61$ \item HFLAV, \href{https://link.springer.com/article/10.1140/epjc/s10052-017-5058-4}{\scriptsize\texttt{Eur. Phys. J. C77 (2017) 895}} \end{itemize}

\end{itemize} \end{block} }

::: :::

Testing LFU with \RDX

\tightmargin ::: columns ::: {.column width=50%}

\small

$\RDX \equiv \frac{\BFDTau}{\BFDMu}$
- Advantageous over measuring BF: cancellation of th. and ex. uncert.
- Precise predictions (1--2%):
  - $\RD = 0.298 \pm 0.004$
  - $\RDst = 0.254 \pm 0.005$
- First anomaly reported in 2012 (BaBar)
- LHCb run 1 measurement in 2022

{ width=90% }

::: ::: {.column width=50%}

\only<2>{ \small

\begin{itemize} \tightlist \item \textbf{Matrix element} of $B \rightarrow D l \neul$ processes factorizable: \end{itemize}

\vspace{-1.5\baselineskip} \tiny \begin{equation*} \mathcal{M}^{\lambda_l}{\lambda_D}(q^2, \theta_l) = \frac{G_F}{\sqrt{2}} \frac{m^2_W}{m^2_W - q^2} \sum{\lambda_W} \eta_{\lambda_W} \textcolor{Red}{L^{\lambda_l}{\lambda_W}(q^2, \theta_l)} \textcolor{Green}{H^{\lambda_D}{\lambda_W}(q^2)} \end{equation*} \vspace{-2\baselineskip}\small

\begin{itemize} \tightlist \item \textbf{\textcolor{Red}{Leptonic currents}} analytically calculable \item \textbf{\textcolor{Green}{Hadronic currents}} involves non-perturbative QCD $\rightarrow$ can't calculate exactly \begin{itemize} \tightlist \item \textbf{Expressed as form factors (FFs), parameterized & constrained} based on \tightlist \begin{itemize} \tightlist \vspace{-\baselineskip} \item Dispersion relations (first principle) \item Heavy Quark Effective Theory (HQET) \end{itemize} \item Numerical values of FF params obtained w/ \textbf{lattice QCD computation & fit to data} \end{itemize} \end{itemize} }

::: :::

About this analysis

::: columns ::: {.column width=60%}

This analysis: measuring \RDX w/ LHCb run 2 (2016--2018) data
- 2016 only for now, but easy to expand
4+ times larger data sample in run 2
- 1.7x intg. lumi. (3.1 \ifb $\rightarrow$ 5.4 \ifb)
- 1.8x prod. xsec. (7 TeV $\rightarrow$ 13 TeV)
- More efficient triggers
- 1.26x sig-like events for 2016 alone (1,734,133 $\rightarrow$ 2,178,793)

::: ::: {.column width=40%} ::: :::

The Large Hadron Collider (LHC)

::: columns ::: {.column width=70%} ::: ::: {.column width=30%}

\tightmargin\small

The LHC is a circular collider
- Circumference: 27 km
- Mainly colliding $pp$
- Run 2 center of mass energy: $\sqrt{s} = 13$ TeV

::: :::

The LHCb detector

::: columns ::: {.column width=45%} \vspace{0.5em} { width=115% } ::: ::: {.column width=55%} ::: :::

::: columns ::: {.column width=70%}

\tightmargin \vspace{0.5em}

LHCb: forward-only spectrometer covering $1.9 < \eta < 4.9$
- $B$ meson produced from gluon fusion $\rightarrow$ $B$ highly boosted
- 4% solid angle coverage, capture ~20% of $\bbbar$
Important for this analysis: tracking & particle identification (PID)
- RICH allow separation of \kaon, \pion, $p$

::: ::: {.column width=30%} \vspace{-4em} { width=120% } ::: :::

Event selection

::: columns ::: {.column width=50%}

Preliminary measurement of \RDX
- Introduction
- Event selection \color{gray}
- Trigger emulation for MC
- Data/MC corrections
- Fit
- Systematics (WIP)

::: ::: {.column width=50%}

Upgrade of the LHCb detector
- \color{gray}Overview of the LHCb upgrade
- The Upstream Tracker

::: :::

Signal and normalization {.fragile}

::: columns ::: {.column width=90%}

Final visible particles (marked in \textcolor{red}{red}): \textcolor{red}{$D^{(*)}\mu$} ::: ::: {.column width=10%} ::: :::

\centering { width=80% }

\visible<2>{ \begin{columns}[T] \begin{column}{0.65\textwidth} \begin{itemize} \tightlist \item (\Bzb \rightarrow \Dstarp (\rightarrow \Dz (\rightarrow \textcolor{red}{\Km \pip})\textcolor{red}{\pip}) \taum (\rightarrow \textcolor{red}{\mun} \neumb \neut) \neutb)) \item (\Bm \rightarrow \Dz (\rightarrow \textcolor{red}{\Km \pip}) \taum (\rightarrow \textcolor{red}{\mun} \neumb \neut) \neutb)) \vspace{0.4\baselineskip} \item (\Bzb \rightarrow \Dstarp (\rightarrow \Dz (\rightarrow \textcolor{red}{\Km \pip})\textcolor{red}{\pip}) \textcolor{red}{\mun} \neumb) \item (\Bm \rightarrow \Dz (\rightarrow \textcolor{red}{\Km \pip}) \textcolor{red}{\mun} \neumb) \end{itemize} \end{column}

\begin{column}{0.35\textwidth} \vspace{6pt}

\RDst, sig

\RD\hspace{1.2pt}, sig

\vspace{0.4\baselineskip} \RDst, norm

\RD\hspace{1.2pt}, norm \end{column} \end{columns} }

\begin{tikzpicture}[relative to page] \node[anchor=north west, execute at begin node=\setlength{\baselineskip}{7pt}, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, ] at (page cs:-0.86, 0.56) { \small $\RDX \equiv \frac{\BFDTau}{\BFDMu} = \frac{\text{BF(sig)}}{\text{BF(norm)}}$ };

\node (lineW) at (page cs:-0.95, -0.64) {};
\node (lineE) at (page cs:0.95, -0.64) {};
\draw<2->[ultra thick,PepsiRed] (lineW) -- (lineE);

\end{tikzpicture}

Selection of \Dz\muon and \Dstarp\muon

::: columns ::: {.column width=50%}

\tightmargin \setlength{\leftmargini}{12pt}

\Dz\muon (\Dz channel)

\begin{itemize} \item \Dz ((\Km\pip) pair)

\begin{itemize} \tightlist \item High (p_T) signature (L0Hadron trigger) \item Invariant mass around \Dz ref. mass \item Displaced from (pp) vertex \end{itemize} \item \muon

\begin{itemize} \tightlist \item No trigger requirement (high \pt-bias) \item PID: add. \UBDT to further reject misID while keeping eff. flat in (p_T) \end{itemize} \end{itemize}

\Dstarp\muon (\Dstar channel)

\Dstarp ($\Dz\pi^+_\text{slow}$ pair)
- Same as \Dz, plus add. low-\pt $\pi^+_\text{slow}$ forming a vertex w/ \Dz

::: ::: {.column width=50%}

\vspace{1em} \resizebox{0.8\textwidth}{!}{ \begin{tikzpicture}[ particle/.style={draw, ->, >=stealth, thick}, vertex/.style={draw, circle, minimum size=9pt, fill=white, inner sep=0pt}, final ptl/.style={inner sep=1pt}, ] \node (a0) at (0, 0) {};

\coordinate[right=2.5em of a0] (a1);
\coordinate[above=2em of a1] (b1);

\node[left=2em of b1, gray, final ptl] (d1) {\tiny $\overline\nu_\tau$};
\coordinate[above right=1.5em and 1.5em of b1] (d2);
\coordinate[below right=0.5em and 2.5em of b1] (d3);

\node[above left=0.5em and 1em of d2, gray, final ptl] (e1) {\tiny $\overline\nu_\mu$};
\node[above right=0.7em and 0.4em of d2, gray, final ptl] (e2) {\tiny $\nu_\tau$};
\node[above right=0.1em and 2.3em of d2, red, final ptl] (e3) {\tiny $\mu^-$};

\node[above right=0.4em and 1.4em of d3, red, final ptl] (f1) {\tiny $K^-$};
\node[below right=1.2em and 0.6em of d3, red, final ptl] (f2) {\tiny $\pi^+$};

\node[below right=1.5em and 0.8em of b1, orange, final ptl] (g1) {\tiny $\pi^+_\text{slow}$};

\draw[particle, dashed] (a0) -- (b1);

\draw[particle, gray] (b1) -- (d1);
\draw[particle, dashed] (b1) -- (d2);
\draw[particle, dashed] (b1) -- (d3);

\draw[particle, gray] (d2) -- (e1);
\draw[particle, gray] (d2) -- (e2);
\draw[particle, red] (d2) -- (e3);

\draw[particle, red] (d3) -- (f1);
\draw[particle, red] (d3) -- (f2);

\draw[particle, orange] (b1) -- (g1);

\node[vertex] (x0) at (a0) {\tiny $pp$};
\node[vertex] (x1) at (b1) {\tiny $B$};
\node[vertex] (x2) at (d3) {\tiny \Dz};
\node[vertex, gray, fill=white] (x3) at (d2) {\tiny $\tau$};

\end{tikzpicture} }

\vspace{0.5em}

\resizebox{0.8\textwidth}{!}{ \begin{tikzpicture}[ particle/.style={draw, ->, >=stealth, thick}, vertex/.style={draw, circle, minimum size=9pt, fill=white, inner sep=0pt}, final ptl/.style={inner sep=1pt}, ] \node (a0) at (0, 0) {};

\coordinate[right=2.5em of a0] (a1);
\coordinate[above=2em of a1] (b1);

\node[left=2em of b1, gray, final ptl] (d1) {\tiny $\overline\nu_\mu$};
\node[above right=1.5em and 1.5em of b1, final ptl, red] (d2) {\tiny $\mu^-$};
\coordinate[below right=0.5em and 2.5em of b1] (d3);

\node[above right=0.4em and 1.4em of d3, red, final ptl] (f1) {\tiny $K^-$};
\node[below right=1.2em and 0.6em of d3, red, final ptl] (f2) {\tiny $\pi^+$};

\node[below right=1.5em and 0.8em of b1, orange, final ptl] (g1) {\tiny $\pi^+_\text{slow}$};

\draw[particle, dashed] (a0) -- (b1);

\draw[particle, gray] (b1) -- (d1);
\draw[particle, red] (b1) -- (d2);
\draw[particle, dashed] (b1) -- (d3);

\draw[particle, red] (d3) -- (f1);
\draw[particle, red] (d3) -- (f2);

\draw[particle, orange] (b1) -- (g1);

\node[vertex] (x0) at (a0) {\tiny $pp$};
\node[vertex] (x1) at (b1) {\tiny $B$};
\node[vertex] (x2) at (d3) {\tiny \Dz};

\end{tikzpicture} }

::: :::

\begin{tikzpicture}[relative to page] \node[anchor=north west, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, text width=3em, align=center ] at (page cs:0.75, 0.4) { \footnotesize sig };

\node[anchor=north west,
      draw=PepsiBlueLt,rounded corners,
      fill=PepsiBlueLt,fill opacity=.22,text opacity=1,
      text width=3em, align=center
    ]
    at (page cs:0.75, -0.3) {
        \footnotesize norm
    };

\node[anchor=north,
      draw=PepsiBlueLt,rounded corners,
      fill=PepsiBlueLt,fill opacity=.22,text opacity=1,
    ]
    at (page cs:0.49, -0.74) {
        \footnotesize \muon from sig decays are softer than \muon from norm
    };

\end{tikzpicture}

\UBDT

::: columns ::: {.column width=50%}

\centering { width=90% }

::: ::: {.column width=50%}

\centering { width=90% }

::: :::

More efficient at rejecting \pion (main source of misID)
Efficiency flat in \pt: sig & norm have similar selection eff. $\rightarrow$ no bias in selection & easier to model
Remaining misID effect (non-\muon misID'ed as \muon) modeled w/ a data ctrl sample

Feed down from \Dstarp\muon and \Dstarz\muon

\small

::: columns ::: {.column width=60%}

\Dstarp\muon: $\Bzb \rightarrow \Dstarp (\rightarrow \Dz\pip) l^- \neulb$

Not all slow \pip from \Dstarp decays are reco'ed
- Some \Dstarp\muon feed down into \Dz\muon

::: {.block}

\Dstarz\muon: $\Bm \rightarrow \Dstarz (\rightarrow \Dz\piz) l^- \neulb$

Neutral slow $\piz$ entirely missed
- All \Dstarz\muon feed down into \Dz\muon
- ~2.5x BF compared to $B \rightarrow \Dz$ :::

::: ::: {.column width=40%}

$p = 8$ GeV for a typical $\pi_\text{slow}$
Fail to reco. ~35% of the time

\centering { width=100% }

::: :::

\vspace{0.5em} ::: columns ::: {.column width=90%}

Feed down makes \RD and $\RDst$ correlated
- Simultaneous fit needed
- Improve precision for \RDst due to large \Dstarz\muon feed down sample

::: ::: {.column width=10%}

::: :::

Background contributions

\tightmargin ::: columns ::: {.column width=50%}

\begin{itemize} \tightlist \item \textbf{Partially reco'ed bkgs} (final states w/ \DXmu + more)

\begin{itemize} \item Four (1P) \Dstst

\begin{itemize}
\tightlist
\item
  \(B \rightarrow \Dstst (\rightarrow D^{0|*|**} (\rightarrow D^{0|*}\pi) \pi) l\neul\)
\end{itemize}

\item<2-> Highly excited \Dstst (\Dstst heavy, (\Dstst_H)):

\begin{itemize}
\tightlist
\item
  \(B \rightarrow \Dstst_H (\rightarrow D^{0|*} \pi\pi) \mu\neum\)
\end{itemize}

\item<3-> \DststS

\begin{itemize}
\tightlist
\item
  \(B_s \rightarrow (D'_{s1}|D_{s2}) (\rightarrow D^{(*)}K) l\neul\)
\end{itemize}

\item<4-> Double-charm backgrounds ((DDX))

\begin{itemize}
\tightlist
\item
  \(B \rightarrow D^{(*)} D_q X\)
\item
  \(D_q \rightarrow \tauon\neut\) when \(q = s\)
\item
  \(D_q \rightarrow K \mu\neum\) when \(q = u \text{ or } d\)
\end{itemize}

\end{itemize} \end{itemize}

::: ::: {.column width=50%}

\visible<5->{ \begin{itemize} \tightlist \item \textbf{Mis-reconstructions}

\begin{itemize} \tightlist \item MisID

\begin{itemize}
\tightlist
\item
  ``\muon'' in the \DXmu pair is \textbf{not} a \muon
\end{itemize}

\item Combinatorial bkgs

\begin{itemize}
\tightlist
\item
  Random combinations of \Dz\muon, \Dstarp\muon, or \Dz\pion
  \textbf{not} from the same \(B\)
\end{itemize}

\end{itemize} \end{itemize} }

\vspace{3em}

\centering

::: :::

\begin{tikzpicture}[relative to page] \node<4->[anchor=north, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, text width=12em, align=center, ] at (page cs:-0.5, -0.54) { \footnotesize\bfseries Modeled w/ MC, w/ shape corrections from ctrl samples }; \node<5->[anchor=north, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, ] at (page cs:0.5, 0.09) { \footnotesize\bfseries Modeled w/ data ctrl samples };

\node[inner sep=0pt] (nw) at (page cs:0.02,-0.1) {};
\node (se) at (page cs:0.92,-0.95) {};

\node (pt) at (page cs:-0.1,0.36) {};

\draw[PepsiBlueLt,ultra thick] (nw) rectangle (se);
\draw[PepsiBlueLt,->,>=stealth,ultra thick] (nw) -- (pt);

\end{tikzpicture}

The isolation BDT

\vspace{4em}

::: columns ::: {.column width=33%}

\resizebox{\textwidth}{!}{ \begin{tikzpicture}[ particle/.style={draw, ->, >=stealth, thick}, vertex/.style={draw, circle, minimum size=9pt, fill=white, inner sep=0pt}, final ptl/.style={inner sep=1pt}, ] \node (a0) at (0, 0) {};

\coordinate[right=2.5em of a0] (a1);
\coordinate[above=2em of a1] (b1);

\node[left=2em of b1, gray, final ptl] (d1) {\tiny $\overline\nu_\mu$};
\node[above right=1.5em and 1.5em of b1, final ptl, red] (d2) {\tiny $\mu^-$};
\coordinate[below right=0.5em and 2.5em of b1] (d3);

\node[above right=0.4em and 1.4em of d3, red, final ptl] (f1) {\tiny $K^-$};
\node[below right=1.2em and 0.6em of d3, red, final ptl] (f2) {\tiny $\pi^+$};

\node[below right=1.5em and 0.8em of b1, orange, final ptl] (g1) {\tiny $\pi^+_\text{slow}$};
\node[above left=1.8em and 2.2em of b1, blue, final ptl] (g2) {\tiny $\pi^\pm$};

\draw[particle, dashed] (a0) -- (b1);

\draw[particle, gray] (b1) -- (d1);
\draw[particle, red] (b1) -- (d2);
\draw[particle, dashed] (b1) -- (d3);

\draw[particle, red] (d3) -- (f1);
\draw[particle, red] (d3) -- (f2);

\draw[particle, orange] (b1) -- (g1);
\draw[particle, blue, dashed] (b1) -- (g2);

\node[vertex] (x0) at (a0) {\tiny $pp$};
\node[vertex] (x1) at (b1) {\tiny $B/\Dstst$};
\node[vertex] (x2) at (d3) {\tiny \Dz};

\end{tikzpicture} }

::: ::: {.column width=33%}

\footnotesize

BDT training variables:
- PV \ipChiSq
- SV \ipChiSq
- track \pt
- track opening
- track \anyChiSq{FD}
- track $\Delta\anyChiSq{FD}$

::: :::

Further divide selected \DXmu samples into sub-samples (skims)
Reject partially reco'ed bkgs with \textcolor{blue}{add. charged track(s)} $\rightarrow$ signal skim
Inverting the selection $\rightarrow$ control skims enriched in such bkgs

\begin{tikzpicture}[relative to page] \node[anchor=south, execute at begin node=\setlength{\baselineskip}{7pt}, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, text width=10em ] at (page cs:0.0, 0.45) { \tiny \bfseries{MVA dist. for $B \rightarrow \Dstst \mu\neum$ bkg (hatched) vs. \textcolor{magenta}{signal (solid)}} };

\node[anchor=south west,
      execute at begin node=\setlength{\baselineskip}{7pt},
      draw=PepsiBlueLt,rounded corners,
      fill=PepsiBlueLt,fill opacity=.22,text opacity=1,
      text width=10em
    ]
    at (page cs:-0.9, 0.45) {
        \tiny
        \bfseries{Schematic $B \rightarrow \Dstst \mu\neum$ decay}
    };

\end{tikzpicture}

Signal and control skims (sub-samples)

\begin{tikzpicture}[relative to page] % ISO \node (isoNW) at (page cs:-0.96,0.75) {}; \node (isoSE) at (page cs:-0.01,-0.08) {};

\draw[normSig,ultra thick] (isoNW) rectangle (isoSE);
\node[anchor=north west,fill=normSig,text=white] (isoTitle) at (isoNW) {\bfseries ISO};
\node[anchor=north west,below right=1pt and -27pt of isoTitle,text width=6em] (isoText) {
    \footnotesize
    Signal-enriched.
    No charged track likely coming from the same $B$ (isolated)
};
\node[anchor=north east,inner sep=0pt] at (page cs:-0.03,0.73) {
    \includegraphics[width=0.33\textwidth]{./section/figs-fit-fit-results/sig-fit/stacked/fit_result-stacked-D0-iso-mmiss2.pdf}
};

% 2OS
\node (2osNW) at (page cs:-0.96,-0.13) {};
\node (2osSE) at (page cs:-0.01,-0.96) {};

\draw[DststH,ultra thick] (2osNW) rectangle (2osSE);
\node[anchor=north west,fill=DststH,text=black] (2osTitle) at (2osNW) {\bfseries 2OS};
\node[anchor=north west,below right=1pt and -28pt of 2osTitle,text width=6em] (2osText) {
    \footnotesize
    Enriched in $B \rightarrow \Dstst_H \mu\neum$.
    Two anti-isolated \pion.
};
\node[anchor=north east,inner sep=0pt] at (page cs:-0.03,-0.15) {
    \includegraphics[width=0.33\textwidth]{./section/figs-fit-fit-results/ctrl-fit/stacked/fit_result-stacked-D0-2os-mmiss2.pdf}
};

% 1OS
\node (1osNW) at (page cs:0.01,0.75) {};
\node (1osSE) at (page cs:0.96,-0.08) {};

\draw[Dstst,ultra thick] (1osNW) rectangle (1osSE);
\node[anchor=north west,fill=Dstst,text=white] (1osTitle) at (1osNW) {\bfseries 1OS};
\node[anchor=north west,below right=1pt and -27pt of 1osTitle,text width=6em] (1osText) {
    \footnotesize
    Enriched in $B \rightarrow \Dstst l\neul$.
    One extra \pion (anti-isolated).
};
\node[anchor=north east,inner sep=0pt] at (page cs:0.94,0.73) {
    \includegraphics[width=0.33\textwidth]{./section/figs-fit-fit-results/ctrl-fit/stacked/fit_result-stacked-D0-1os-mmiss2.pdf}
};

% DD
\node (ddNW) at (page cs:0.01,-0.13) {};
\node (ddSE) at (page cs:0.96,-0.96) {};

\draw[DD,ultra thick] (ddNW) rectangle (ddSE);
\node[anchor=north west,fill=DD,text=black] (ddTitle) at (ddNW) {\bfseries DD};
\node[anchor=north west,below right=1pt and -24pt of ddTitle,text width=6em] (ddText) {
    \footnotesize
    Enriched in $B \rightarrow D^{(*)} D_q X$.
    One or more anti-isolated tracks, at least one \kaon
};
\node[anchor=north east,inner sep=0pt] at (page cs:0.94,-0.15) {
    \includegraphics[width=0.33\textwidth]{./section/figs-fit-fit-results/ctrl-fit/stacked/fit_result-stacked-D0-dd-mmiss2.pdf}
};

\end{tikzpicture}

Key kinematic variables: \mmSq, \el, \qSq

\small

Take advantage of \mmSq, \el, \qSq to separate sig, norm, and bkgs
- $\mmSq \equiv (p_B - p_{D^{(*)}} - p_l)^2$
- $\el$: lepton energy in $B$ rest frame
- $q^2 \equiv (p_B - p_{D^{(*)}})^2$

\vspace{-0.8em} ::: columns ::: {.column width=33%} ::: ::: {.column width=33%} ::: ::: {.column width=33%} ::: :::

\vspace{2.5em}

Not known exactly in hadron colliders ($B$ momenta not known exactly)
- Can be approximated with rest frame approximation (RFA)

\begin{tikzpicture}[relative to page] \node[anchor=north west, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1 ] at (page cs:0.02, 0.575) {These are fit variables!};

\node[anchor=north west,
      draw=PepsiBlueLt,rounded corners,
      fill=PepsiBlueLt,fill opacity=.22,text opacity=1
    ]
    at (page cs:-0.515, 0.30) {\footnotesize \mmSq};

\node[anchor=north west,
      draw=PepsiBlueLt,rounded corners,
      fill=PepsiBlueLt,fill opacity=.22,text opacity=1
    ]
    at (page cs:0.175, 0.30) {\footnotesize \el};

\node[anchor=north west,
      draw=PepsiBlueLt,rounded corners,
      fill=PepsiBlueLt,fill opacity=.22,text opacity=1
    ]
    at (page cs:0.41, 0.30) {\footnotesize \qSq};

\node[anchor=north,
      draw=PepsiBlueLt,rounded corners,
      fill=PepsiBlueLt,fill opacity=.22,text opacity=1,
      text width=30em
    ]
    at (page cs:00, -0.44) {
        \footnotesize
        $\textcolor{blue}{\Bm \rightarrow \Dz\taum\neutb}$
        vs
        $\textcolor{red}{\Bm \rightarrow \Dz\mun\neumb}$
        vs
        $\textcolor{orange}{\Bzb \rightarrow \Dstarp\mun\neumb}$ feed down
        vs
        $\textcolor{gray}{\Bm \rightarrow \Dstarz\mun\neumb}$ feed down
        (RFA)
    };

\end{tikzpicture}

Rest frame approximation

\tightmargin\vspace{-0.5em} ::: columns ::: {.column width=50%}

\small

$e^- e^+$ collider
- $\sqrt{s}$ at $\Upsilon(4S)$ resonance (~10.58 GeV)
- $B$ meson production: $e^- e^+ \rightarrow \Upsilon(4S) \rightarrow B \Bbar$
- Tag fully reco'ed $B$ meson, $B_\text{tag}$
- $p_{B_\text{sig}} = p_{e^-} + p_{e^+} - p_{B_\text{tag}}$

\resizebox{\textwidth}{!}{ \begin{tikzpicture}[particle/.style={draw, ->, >=stealth, thick}] \node (a0) at (0, 0) {}; \node[right=2.5em of a0] (a1) {$e^+$}; \node[left=2.5em of a0] (a2) {$e^-$};

\coordinate[above=2em of a1] (b1);
\coordinate[below=2em of a2] (b2);

\coordinate[below left=0.5em and 2em of b2] (c1);
\node[below left=2em and 1em of b2, blue] (c2) {hadronic particles};
\coordinate[below left=2em and 1em of b2] (c3);

\node[left=2em of b1, gray] (d1) {$\overline\nu_\tau$};
\coordinate[above right=1.5em and 1.5em of b1] (d2);
\coordinate[below right=0.5em and 2.5em of b1] (d3);

\node[above left=0.5em and 1em of d2, gray] (e1) {$\overline\nu_l$};
\node[above right=0.7em and 0.4em of d2, gray] (e2) {$\nu_\tau$};
\node[above right=0.1em and 2.3em of d2, red] (e3) {$l^-$};

\coordinate[above right=0.2em and 1.2em of d3] (f1);
\coordinate[below right=1em and 0.4em of d3] (f3);

\draw [particle] (a1) -- (a0);
\draw [particle] (a2) -- (a0);

\draw [particle, dashed, red] (a0) -- (b1) node[midway, left, xshift=-5pt] {$B_\text{sig}$};
\draw [particle, dashed, blue] (a0) -- (b2) node[midway, left, xshift=-5pt] {$B_\text{tag}$};

\draw [particle, blue] (b2) -- (c1);
\draw [particle, blue] (b2) -- (c2);
\draw [particle, blue] (b2) -- (c3);

\draw [particle, gray] (b1) -- (d1);
\draw [particle, red, dashed] (b1) -- (d2) node[midway, left] {$\tau^-$};
\draw [particle, red, dashed] (b1) -- (d3) node[midway, below] {$D^0$};

\draw[particle, gray] (d2) -- (e1);
\draw[particle, gray] (d2) -- (e2);
\draw[particle, red] (d2) -- (e3);

\draw[particle, red] (d3) -- (f1);
\draw[particle, red] (d3) -- (f3);

\end{tikzpicture} }

::: ::: {.column width=50%}

\small

$pp$ collider
- $\sqrt{s} \gg \Upsilon(4S)$ resonance (13 TeV)
- $B$ meson production: $\text{partons} \rightarrow \bbbar \rightarrow B\Bbar$
  - $p_\text{partons}$ unknown
- $B$ vertex known to high precision
  - Visible part of $B$: $m_\text{vis}$, $p_\text{vis}$
  - Angle between $B$ flight dir & $z$ axis: $\alpha$
  - Assume: proper velocity ($\gamma\beta$) the same in $z$ for $B$ and vis $\rightarrow$ $(p_B)z = \frac{m_B}{m\text{vis}}(p_\text{vis})_z$
- RFA: $|p_B| = \frac{m_B}{m_\text{vis}} (p_\text{vis})_z \sqrt{1 + \tan^2\alpha}$

\resizebox{0.75\textwidth}{!}{ \begin{tikzpicture}[particle/.style={draw, ->, >=stealth, thick}] \node (a0) at (0, 0) {}; \node[right=2.5em of a0] (a1) {$p$ ($z$ axis)}; \node[left=2.5em of a0] (a2) {$p$};

\coordinate[above=2em of a1] (b1);
\coordinate[below=2em of a2] (b2);
\coordinate[below=0.9em of b1] (b3);

\node[left=2em of b1, gray] (d1) {$\overline\nu_\tau$};
\coordinate[above right=1.5em and 1.5em of b1] (d2);
\coordinate[below right=0.5em and 2.5em of b1] (d3);

\node[above left=0.5em and 1em of d2, gray] (e1) {$\overline\nu_l$};
\node[above right=0.7em and 0.4em of d2, gray] (e2) {$\nu_\tau$};
\node[above right=0.1em and 2.3em of d2, red] (e3) {$l^-$};

\coordinate[above right=0.2em and 1.2em of d3] (f1);
\coordinate[below right=1em and 0.4em of d3] (f3);

\draw [particle] (a1) -- (a0);
\draw [particle] (a2) -- (a0);

\draw [particle, orange] (a0) -- (b1) node[midway, left, xshift=-5pt] {$B$ flight dir};
\draw [particle, red] (a0) -- (b3) node[midway, right, xshift=8pt] {visible};

\draw [particle, gray] (b1) -- (d1);
\draw [particle, red, dashed] (b1) -- (d2) node[midway, left] {$\tau^-$};
\draw [particle, red, dashed] (b1) -- (d3) node[midway, above, xshift=6pt] {$D^0$};

\draw[particle, gray] (d2) -- (e1);
\draw[particle, gray] (d2) -- (e2);
\draw[particle, red] (d2) -- (e3);

\draw[particle, red] (d3) -- (f1);
\draw[particle, red] (d3) -- (f3);

\draw pic["\tiny$\textcolor{blue}{\alpha}$",
          draw=blue,thick,-,angle eccentricity=1.3,angle radius=18pt,
          fill=blue,fill opacity=.5,text opacity=1]
    {angle=a1--a0--b1};

\end{tikzpicture} }

::: :::

\begin{tikzpicture}[relative to page] \node[anchor=north west, execute at begin node=\setlength{\baselineskip}{7pt}, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, ] at (page cs:-0.9, 0.06) { \scriptsize No RFA needed at $B$ factories };

\node (recNW) at (page cs:-0.99, 0.78) {};
\node (recSE) at (page cs:-0.02, -0.98) {};
\draw<2>[fill=white,white] (recNW) rectangle (recSE);

\node<2>[anchor=north,inner sep=0pt] at (page cs: -0.49,0.65) {
    \includegraphics[width=20em]{./slides-figures/rfa_resolution.pdf}
};

\node<2>[anchor=north,
      execute at begin node=\setlength{\baselineskip}{7pt},
      draw=PepsiBlueLt,rounded corners,
      fill=PepsiBlueLt,fill opacity=.22,text opacity=1,
      text width=18em,
    ]
    at (page cs:-0.46, -0.53) {
        \small LHCb resolution worse than $B$ factories,
        RFA resolution comparable to LHCb's
    };

\end{tikzpicture}

Trigger emulation for MC

::: columns ::: {.column width=50%}

Preliminary measurement of \RDX
- Introduction
- Event selection
- Trigger emulation for MC \color{gray}
- Data/MC corrections
- Fit
- Systematics (WIP)

::: ::: {.column width=50%}

Upgrade of the LHCb detector
- \color{gray}Overview of the LHCb upgrade
- The Upstream Tracker

::: :::

Tracker-only MC

::: columns ::: {.column width=50%}

\tightmargin

Leading sys. uncert. in run 1: MC stats
Run 2: ~4x more data $\rightarrow$ need even more MC
- Computationally impractical to simulate all detector responses
- ~85% computation time spent on RICH and calorimeters $\rightarrow$ ~8x faster turning them off
- Requested ~7.3B MC (run 1: ~1B), w/ ~1,679M on disk (run 1: ~65M)
  - Huge challenge logistically
Use Tracker-only (TO) MC $\rightarrow$ only tracking system turned on
- Triggers rely on calorimeters $\rightarrow$ emulate trigger offline

::: ::: {.column width=50%} \vspace{2em} ::: :::

\begin{tikzpicture}[relative to page] \node[anchor=north, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, ] at (page cs:0.46, 0.7) { \tiny \RDX run 1 analysis uncertainty table }; \end{tikzpicture}

Emulate L0

::: columns ::: {.column width=50%} ::: ::: {.column width=50%} ::: :::

\tightmargin ::: columns ::: {.column width=50%}

\small

L0Hadron TOS
- Trained a BDT (\xgboost) to predict the trigger probabilistically
- Based on tracker estimated $E_T$, calo-hitting projections & more

::: ::: {.column width=50%}

\small

L0Global TIS
- Measured in data ($B \rightarrow \jpsi K$) b.c. L0Global TIS portable across reco modes, applied as a weight

::: :::

Emulate HLT

::: columns ::: {.column width=50%} ::: ::: {.column width=50%} ::: :::

\tightmargin ::: columns ::: {.column width=50%}

\small

Hlt1TrackMVA
- Relevant vars exist in TO MC
- ~1% constant diff after applying online/offline correction
- Further corrected in final reweighting

::: ::: {.column width=50%}

\small

Hlt1TwoTrackMVA
- Similarly processed as Hlt1TrackMVA
- ~2.3% constant diff after online/offline correction

::: :::

Data/MC corrections

::: columns ::: {.column width=50%}

Preliminary measurement of \RDX
- Introduction
- Event selection
- Trigger emulation for MC
- Data/MC corrections \color{gray}
- Fit
- Systematics (WIP)

::: ::: {.column width=50%}

Upgrade of the LHCb detector
- \color{gray}Overview of the LHCb upgrade
- The Upstream Tracker

::: :::

Procedure overview

\tightmargin\small @. Apply known corrections (initial reweighting) - \textcolor{PepsiBlueLt}{Tracking efficiency} - \textcolor{PepsiRed}{$B$ kinematic and multiplicity}

\vspace{11em} @. Update MC FF parameterizations for \Dz, \Dstar, \Dstst @. Perform a fit to estimate yields of sig, norm, & bkgs @. Correct add. kinematic & geometric vars in low-\mmSq region (final reweighting) - Enriched in norm decays ($B \rightarrow D^{(*)}\mu\neum$) - Diff. in that region due to modelling of detector effects

\begin{tikzpicture}[relative to page] % Tracking \node (trkNW) at (page cs:-0.91,0.44) {}; \node (trkSE) at (page cs:-0.02,-0.46) {}; \draw[PepsiBlueLt,ultra thick] (trkNW) rectangle (trkSE) node[pos=.5] (trkCtn) {};

\node[xshift=-2pt,yshift=-7pt] (trkFig) at (trkCtn) {
    \includegraphics[width=0.44\textwidth]{./chapter/figs-mc-correction/reweighting-tracking/tracking_eff_2016.pdf}
};
% Remarks
\node[draw=PepsiBlueLt,rounded corners,
      fill=PepsiBlueLt,fill opacity=.22,text opacity=1,
    ]
    at (trkFig) {
    \tiny $\epsilon(\text{tracking, data}) / \epsilon(\text{tracking, MC})$
    };
% Title
\node[anchor=north west,fill=PepsiBlueLt,text=white,inner sep=3pt] (trkTitle) at (trkNW) {
    \scriptsize\bfseries Tracking efficiency
};
\draw[PepsiBlueLt,ultra thick] (trkNW) rectangle (trkSE);

% B prod kinematics
\node (prodNW) at (page cs:0.02,0.44) {};
\node (prodSE) at (page cs:0.91,-0.46) {};
\draw[PepsiRed,ultra thick] (prodNW) rectangle (prodSE) node[pos=.5] (prodCtn) {};

% Draw bottom first so top can cover bot titles
\node[xshift=-5.3em,yshift=-3.1em] (pPt) at (prodCtn) {
    \includegraphics[width=0.18\textwidth]{./chapter/figs-mc-correction/reweighting-JpsiK/reweight-JpsiK/b_pt.pdf}
};
\node[xshift=4.8em,yshift=-3.1em] (pEta) at (prodCtn) {
    \includegraphics[width=0.18\textwidth]{./chapter/figs-mc-correction/reweighting-JpsiK/reweight-JpsiK/b_eta.pdf}
};
% Top plots
\node[xshift=-5.3em,yshift=2em] (pNtracks) at (prodCtn) {
    \includegraphics[width=0.18\textwidth]{./chapter/figs-mc-correction/reweighting-JpsiK/reweight-JpsiK/ntracks.pdf}
};
\node[xshift=4.8em,yshift=2em] (pNdof) at (prodCtn) {
    \includegraphics[width=0.18\textwidth]{./chapter/figs-mc-correction/reweighting-JpsiK/reweight-JpsiK/b_ownpv_ndof.pdf}
};


% Remarks
\node[draw=PepsiRed,rounded corners,
      fill=PepsiRed,fill opacity=.22,text opacity=1,
      inner sep=1pt,xshift=1.8em
    ] at (pNtracks) {\tiny nTracks};
\node[draw=PepsiRed,rounded corners,
      fill=PepsiRed,fill opacity=.22,text opacity=1,
      inner sep=1pt,xshift=1.2em
    ] at (pNdof) {\tiny PV NDOF};
\node[draw=PepsiRed,rounded corners,
      fill=PepsiRed,fill opacity=.22,text opacity=1,
      inner sep=1pt,xshift=1.6em
    ] at (pPt) {\tiny $B$ \pt};
\node[draw=PepsiRed,rounded corners,
      fill=PepsiRed,fill opacity=.22,text opacity=1,
      inner sep=1pt,xshift=2.1em
    ] at (pEta) {\tiny $B$ $\eta$};
% Title
\node[anchor=north west,fill=PepsiRed,text=white,inner sep=3pt] (trkTitle) at (prodNW) {
    \scriptsize\bfseries $B$ kinematics and multiplicity
};
\draw[PepsiRed,ultra thick] (prodNW) rectangle (prodSE);

\end{tikzpicture}

Form factor reweighting

\small\vspace{-1.2em}

Change FF parameterization $\xleftrightarrow{\text{\bfseries equivalent}}$ reweighting
- FF parameterization determines differential decay rate $d\Gamma / d\PSpt$
- For each MC event, weight $w$ given by: \vspace{-0.4\baselineskip} $$ \scriptsize w = \left. \frac{d\Gamma_\text{target} / d\PSpt}{d\Gamma_\text{source} / d\PSpt} \right|_\text{eval at phase space point} $$ \vspace{-1.1\baselineskip}
BGL & BLR more flexible $\rightarrow$ derive shape corrections in fit \vspace{-0.8\baselineskip}

::: columns ::: {.column width=33%} ::: ::: {.column width=33%} ::: ::: {.column width=33%} ::: :::

\begin{tikzpicture}[relative to page] \node[anchor=north, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, inner sep=2pt ] at (page cs:-0.69, -0.46) { \tiny $\Bm \rightarrow \Dz\mun\neumb$ };

\node[anchor=north,
      draw=PepsiBlueLt,rounded corners,
      fill=PepsiBlueLt,fill opacity=.22,text opacity=1,
    ]
    at (page cs:-0.62, -0.78) {
        \footnotesize $B \rightarrow \Dz$ (CLN $\rightarrow$ \textcolor{red}{BGL})
    };

\node[anchor=north,
      draw=PepsiBlueLt,rounded corners,
      fill=PepsiBlueLt,fill opacity=.22,text opacity=1,
      inner sep=2pt
    ]
    at (page cs:0.02, -0.26) {
        \tiny $\Bzb \rightarrow \Dstarp\mun\neumb$
    };

\node[anchor=north,
      draw=PepsiBlueLt,rounded corners,
      fill=PepsiBlueLt,fill opacity=.22,text opacity=1,
    ]
    at (page cs:0.02, -0.78) {
        \footnotesize $B \rightarrow \Dstar$ (CLN $\rightarrow$ \textcolor{red}{BGL})
    };

\node[anchor=north,
      draw=PepsiBlueLt,rounded corners,
      fill=PepsiBlueLt,fill opacity=.22,text opacity=1,
      inner sep=2pt
    ]
    at (page cs:0.78, -0.26) {
        \tiny $\Bzb \rightarrow D^{*+}_2\mun\neumb$
    };

\node[anchor=north,
      draw=PepsiBlueLt,rounded corners,
      fill=PepsiBlueLt,fill opacity=.22,text opacity=1,
    ]
    at (page cs:0.68, -0.78) {
        \footnotesize $B \rightarrow \Dstst$ (ISGW2 $\rightarrow$ \textcolor{red}{BLR})
    };

\end{tikzpicture}

Final reweighting

\begin{tikzpicture}[relative to page] \node[anchor=north, execute at begin node=\setlength{\baselineskip}{7pt}, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, inner sep=2pt ] at (page cs:0.34, 0.72) { \scriptsize Multi-stage final reweighting vars };

\node[anchor=north,
      execute at begin node=\setlength{\baselineskip}{7pt},
      draw=Red,rounded corners,
      fill=Red,fill opacity=.22,text opacity=1,
      inner sep=2pt
    ]
    at (page cs:-0.015, -0.77) {\scalebox{.55}{S1: $\Dz \sqrt{IP \chi^2}/IP)$}};

\node[anchor=north,
      execute at begin node=\setlength{\baselineskip}{7pt},
      draw=Blue,rounded corners,
      fill=Blue,fill opacity=.22,text opacity=1,
      inner sep=2pt
    ]
    at (page cs:0.36, -0.77) {\scalebox{.55}{S2: $\Dz\muon \log(FD \chi^2)$}};

\node[anchor=north,
      execute at begin node=\setlength{\baselineskip}{7pt},
      draw=Green,rounded corners,
      fill=Green,fill opacity=.22,text opacity=1,
      inner sep=2pt
    ]
    at (page cs:0.745, -0.77) {\scalebox{.55}{S10: $\Dz \log(1-DIRA)$}};

% picture in the left
\node[anchor=north west,inner sep=0pt]
    at (page cs:-1.0, 0.79) {
        \includegraphics[width=15em]{./slides-figures/schematic_final_rwt.pdf}
    };

\end{tikzpicture}

\vspace{-2.5em}\tightmargin ::: columns ::: {.column width=35%}

\vspace{10em}\small

Perform an initial fit to estimate yields
Reweight low \mmSq region of sig (ISO) fit
- Enriched in norm
After final reweighting, consider MC describe data sufficiently well

::: ::: {.column width=65%}

\vspace{2.1em} \resizebox{\textwidth}{!}{ \begin{tabular}{c|l|c|l|c|l} \toprule {\bf Variable 1} & {\bf Binning 1} & {\bf Variable 2} & {\bf Binning 2} & {\bf Variable 3} & {\bf Binning 3} \ \midrule $K$ $\sqrt{IP, \chi^2} / IP$ & 10, 0 -- 100 & $\pi$ $\sqrt{IP, \chi^2} / IP$ & 10, 0 -- 100 & \textcolor{Red}{$D^0$ $\sqrt{IP, \chi^2} / IP$} & 10, 15 -- 110 \ \textcolor{Blue}{$D^0\mu$ $\log(FD, \chi^2)$} & 10, 4 -- 12.5 & $D^0$ $\log(IP, \chi^2)$ & 10, 2 -- 9 & $\mu$ $\log(IP, \chi^2)$ & 10, 3.6 -- 11 \ $K$ $p_T$ [GeV] & 10, 0 -- 11 & $K$ $\log(IP, \chi^2)$ & 10, 3.6 -- 10.2 & $K$ $\sqrt{IP, \chi^2} / IP$ & 10, 5 -- 100 \ $\pi$ $p_T$ [GeV] & 10, 0 -- 12.5 & $\pi$ $\log(IP, \chi^2)$ & 10, 3.6 -- 10.2 & $\pi$ $\sqrt{IP, \chi^2} / IP$ & 10, 5 -- 100 \ $\mu$ $p_T$ [GeV] & 10, 0 -- 12 & $\mu$ $\log(IP, \chi^2)$ & 10, 3.6 -- 10.8 & $\mu$ $\sqrt{IP, \chi^2} / IP$ & 10, 0 -- 100 \ $D^0$ $p_T$ [GeV] & 10, 2 -- 18.5 & $D^0$ $\log(IP, \chi^2)$ & 10, 2 -- 9 & $D^0$ $\sqrt{IP, \chi^2} / IP$ & 10, 18 -- 102 \ $D^0$ $p_T$ [GeV] & 20, 2 -- 18.5 & $D^0$ $\eta$ & 10, 1.8 -- 5 & -- & -- \ $K$ $p_T$ [GeV] & 20, 0 -- 11 & $K$ $\eta$ & 10, 1.8 -- 5 & -- & -- \ $\pi$ $p_T$ [GeV] & 20, 0 -- 12.5 & $\pi$ $\eta$ & 10, 1.8 -- 5 & -- & -- \ $\mu$ $p_T$ [GeV] & 20, 0 -- 12 & $\mu$ $\eta$ & 10, 1.8 -- 5 & -- & -- \ \textcolor{Green}{$D^0$ $\log(1 - DIRA)$} & 20, -14.2 -- -8.4 & -- & -- & -- & -- \ slow $\pi$ $p_T$ [GeV]\parnote{ This is for \Dstar channel only. } & 6, 0 -- 1.6 & slow $\pi$ $\eta$ & 10, 1.8 -- 4.8 & -- & -- \ slow $\pi$ $p_T$ [GeV]\parnoteref{parnote:final-rwt-dst} & 6, 0 -- 1.6 & slow $\pi$ $\log(IP, \chi^2)$ & 10, -4 -- 7 & slow $\pi$ $\sqrt{IP, \chi^2} / IP$ & 10, 0 -- 50 \ \bottomrule \end{tabular} }

\vspace{0.5em}

{ width=32% } { width=32% } { width=32% }

::: :::

Fit

::: columns ::: {.column width=50%}

Preliminary measurement of \RDX
- Introduction
- Event selection
- Trigger emulation for MC
- Data/MC corrections
- Fit \color{gray}
- Systematics (WIP)

::: ::: {.column width=50%}

Upgrade of the LHCb detector
- \color{gray}Overview of the LHCb upgrade
- The Upstream Tracker

::: :::

Overall fit strategy

::: columns ::: {.column width=65%}

\tightmargin \begin{itemize} \tightlist \item \textbf{Binned maximum likelihood fit}

\begin{itemize} \tightlist \item Fit vars: \mmSq, \el, \qSq \item \textbf{Norm, sig, bkgs represented by fit templates}

\begin{itemize}
\tightlist
\item
  3D histos, correlation-preserving
\end{itemize}

\end{itemize} \item<2-> Fit control skims (1OS, 2OS, DD) \textbf{first}

\begin{itemize} \tightlist \item 3 control skims per channel (\rightarrow) \textbf{simultaneous fit to 6 datasets} \item \textbf{Derive shape corrections for bkgs}

\begin{itemize}
\tightlist
\item
  FF variations (5) \& data-driven corrections (12)
\end{itemize}

\end{itemize} \item<3-> Fit signal skim (ISO)

\begin{itemize} \tightlist \item \textbf{Simultaneous fit to \Dz & \Dstar ISO skim} \item \textbf{Load params for bkg shape corrections} as constraints or fully fixed

\begin{itemize}
\tightlist
\item
  \textbf{They can't be determined precisely in sig fit}
\item
  Compared to \textbf{\textcolor{Red}{nominal fit}},
  \textbf{\textcolor{Green}{fit w/ 1 add. $DDX$ param floating}} has
  smaller pulls but drives signal yield to 0
\end{itemize}

\end{itemize} \end{itemize}

::: ::: {.column width=30%}

\visible<3->{ \vspace{1.1em}

\includegraphics{./chapter/figs-fit/fit_uvsd/fit_result-stacked-Dst-iso-q2.pdf}

\includegraphics{./chapter/figs-fit/fit_uvsd/fit_result-stacked-Dst-iso-q2-floating_uvsd.pdf} }

::: :::

\begin{tikzpicture}[relative to page] \node<3->[anchor=north, draw=Red,rounded corners, fill=Red,fill opacity=.22,text opacity=1, inner sep=2pt ] at (page cs:0.66, 0.72) {\tiny Nominal fit for \Dstar ISO};

\node<3->[anchor=north,
      execute at begin node=\setlength{\baselineskip}{6pt},
      draw=Green,rounded corners,
      fill=Green,fill opacity=.22,text opacity=1,
      inner sep=2pt,
      text width=8em, align=center
    ]
    at (page cs:0.66, -0.79) {\tiny \Dstar ISO fit w/ 1 $DDX$ shape param floating};

\end{tikzpicture}

Templates in the fit

\begin{tikzpicture}[relative to page] % main figure \node[anchor=north west,inner sep=0pt] at (page cs:-1, 0.80) { \includegraphics[width=22em]{./slides-figures/fit_templates_schematic.pdf}};

% block pulls
\node (blockPullNW) at (page cs:-1,-0.152) {};
\node (blockPullSE) at (page cs:1,-1) {};
\draw[fill=white,white] (blockPullNW) rectangle (blockPullSE);

% block '5' (the remaining tick label)
\node (block5NW) at (page cs: -1,-0.1) {};
\node (block5SE) at (page cs: -0.84,-1) {};
\draw[fill=white,white] (block5NW) rectangle (block5SE);

% erase title
\node[anchor=north,inner sep=0pt,fill=white,white,text width=8em,align=center] at (page cs:-0.39,0.781) {
    \scriptsize \Dz, DD, \el
};

\node[anchor=north,inner sep=3pt,rounded corners,
      draw=gray,fill=gray,fill opacity=.22,text opacity=1] at (page cs:-0.4,0.6) {
    \footnotesize \Dz, DD, \el
};

\node[anchor=north west] at (page cs: 0.1,0.75) {
    \scriptsize
    \begin{tabular}{l|c|c}
        & \Dz & \Dstar \\
        \midrule
        \colorbox{Ds}{\phantom{XXX}} \DststS & 2 subtypes & 2 subtypes \\
        \colorbox{DD}{\phantom{XXX}} $DDX$ & 4 & 4 \\
        \colorbox{DststH}{\phantom{XXX}} $\Dstst_H$ & 3 & 1 \\
        \colorbox{Dstst}{\phantom{XXX}} \Dstst & 20 & 12 \\
        \colorbox{normSig}{\phantom{XXX}} norm + sig & 6 & 2 \\
        \colorbox{combBkg}{\phantom{XXX}} comb bkg & 1 & 2 \\
        \colorbox{misID}{\phantom{XXX}} misID & 1 & 1 \\
        \midrule
        \colorbox{white}{\phantom{XXX}} Total & 37 & 24 \\
    \end{tabular}
};

\end{tikzpicture}

\vspace{9em} ::: columns ::: {.column width=33%} \centering \vspace{4pt} ::: ::: {.column width=33%} \vspace{3pt} ::: ::: {.column width=33%} \vspace{2pt} ::: :::

\begin{tikzpicture}[relative to page] % misID \node (misIDNW) at (page cs:-0.95,-0.19) {}; \node (misIDSE) at (page cs:-0.32,-0.95) {}; \draw[misID,ultra thick] (misIDNW) rectangle (misIDSE);

\node[execute at begin node=\setlength{\baselineskip}{7pt},
      anchor=north,inner sep=1pt,draw=misID,fill=misID,fill opacity=.4,text opacity=1,
      rounded corners,text width=7em] at (page cs:-0.55,-0.23) {
    \tiny misID: iterative unfolding \\
    \colorbox{normSig}{\phantom{XX}} \pion
    \colorbox{Dstst}{\phantom{XX}} ghost
};

% comb bkg
% block plot title
\node (combNW) at (page cs:-0.29,-0.19) {};
\node (combSE) at (page cs:0.95,-0.215) {};
\draw[fill=white,white] (combNW) rectangle (combSE);

\node (combNW) at (page cs:-0.29,-0.19) {};
\node (combSE) at (page cs:0.95,-0.95) {};
\draw[combBkg,ultra thick] (combNW) rectangle (combSE);

% BComb
\node[execute at begin node=\setlength{\baselineskip}{7pt},
      anchor=north,inner sep=1pt,draw=combBkg,fill=combBkg,fill opacity=.4,text opacity=1,
      rounded corners,text width=8em] at (page cs:0.07,-0.41) {
    \tiny \DXmu comb: lin. rescale fac. (fit to $m_B$ USB)
};

% Dst comb
\node[execute at begin node=\setlength{\baselineskip}{7pt},
      anchor=north,inner sep=1pt,draw=combBkg,fill=combBkg,fill opacity=.4,text opacity=1,
      rounded corners,text width=6em] at (page cs:0.71,-0.60) {
    \tiny $\Dz\pi$ comb: rescale to fitted yield
};

\end{tikzpicture}

Fit model

\vspace{0.5em} ::: columns ::: {.column width=50%}

\resizebox{\textwidth}{!}{ \begin{tabular}{r|c|c|l} \toprule \textbf{Alias} & \textbf{Decay mode} & \textbf{Normalization} & \textbf{Index} \ \midrule \texttt{D_Dmu} & $B^- \rightarrow D^0 \mu^- \overline{\nu}\mu$ & $N{D \mu}$ & 1 \ \texttt{D_dDstmu} & $\overline{B}^0 \rightarrow D^{+} \mu^- \overline{\nu}\mu$ & $N{D \mu} \times r_{D^}^\text{isospin} \times r_{D^{0}}^{0}$ & 2 \ \texttt{D_uDstmu} & $B^- \rightarrow D^{0} \mu^- \overline{\nu}\mu$ & $N{D \mu} \times r_{D^{0}}^{0}$ & 3 \ \texttt{D_Dtau} & $B^- \rightarrow D^0 \tau^- \overline{\nu}\tau$ & $N{D \mu} \times \textcolor{red}{\eta_{D^0}} \times \mathcal{R}(D)$ & 4 \ \texttt{D_dDsttau} & $\overline{B}^0 \rightarrow D^{+} \tau^- \overline{\nu}\tau$ & $N{D \mu} \times \textcolor{red}{\frac{\eta_{D^{+}}}{\tilde{\eta}_{D^{+}}}} \times \textcolor{red}{\tilde{\eta}{D^{+}}} \times \mathcal{R}(D^) \times r{D^{0}}^{0} \times r_{D^}^\text{isospin}$ & 5 \ \texttt{D_uDsttau} & $B^- \rightarrow D^{0} \tau^- \overline{\nu}\tau$ & $N{D \mu} \times \textcolor{red}{\frac{\eta_{D^{0}}}{\tilde{\eta}_{D^{+}}}} \times \textcolor{red}{\tilde{\eta}_{D^{+}}} \times \mathcal{R}(D^) \times r_{D^{0}}^{0}$ & 6 \ \texttt{D_dD1mu} & $\overline{B}^0 \rightarrow D_1 \mu \overline{\nu}\mu$ & $\textcolor{blue}{\rho^\text{isospin}{D_1}} \times \textcolor{red}{n^{0}{D^{**}}} \times N{D \mu} \times \textcolor{red}{f^{0}{D_1}} \times \textcolor{red}{\epsilon{D_1}} \times \mathcal{B}^{0}{D_1}$ & 7 \ \texttt{D_dD1mu_pipi} & $\overline{B}^0 \rightarrow D_1 (\rightarrow D^0 \pi\pi) \mu \overline{\nu}\mu$ & $\textcolor{blue}{\rho^\text{isospin}{D_1\pi\pi}} \times \textcolor{red}{n^{0}{D^{**}}} \times N_{D \mu} \times \textcolor{red}{f^{0}{D_1}} \times \textcolor{red}{\epsilon{D_1\pi\pi}} \times \mathcal{B}^{0}_{D_1\pi\pi}$ & 8 \ \texttt{D_dD2mu} & $\overline{B}^0 \rightarrow D^2 \mu \overline{\nu}\mu$ & $\textcolor{blue}{\rho^\text{isospin}_{D_2^}} \times \textcolor{red}{n^{0}{D^{**}}} \times N{D \mu} \times \textcolor{red}{f^{0}{D_2^*}} \times \textcolor{red}{\epsilon{D_2^}} \times \mathcal{B}^{0}_{D_2^}$ & 9 \ \texttt{D_dD1pmu} & $\overline{B}^0 \rightarrow D'1 \mu \overline{\nu}\mu$ & $\textcolor{blue}{\rho^\text{isospin}{D'1}} \times \textcolor{red}{n^{0}{D^{**}}} \times N{D \mu} \times \textcolor{red}{f^{0}{D'1}} \times \textcolor{red}{\epsilon{D'1}} \times \mathcal{B}^{0}{D'1}$ & 10 \ \texttt{D_dD0mu} & $\overline{B}^0 \rightarrow D^*0 \mu \overline{\nu}\mu$ & $\textcolor{blue}{\rho^\text{isospin}{D_1^*}} \times \textcolor{red}{n^{0}{D^{}}} \times N_{D \mu} \times \textcolor{red}{f^{0}{D_1^*}} \times \textcolor{red}{\epsilon{D_1^}} \times \mathcal{B}^{0}_{D_1^}$ & 11 \ \texttt{D_Dstzpipimu} & $\overline{B} \rightarrow D^{} (\rightarrow D^{0} \pi\pi) \mu \overline{\nu}\mu$ & $\textcolor{red}{n^{0}{D^{}}} \times N_{D \mu} \times \textcolor{red}{f_\text{guess}} \times f^0_{D^{*0}\pi\pi}$ & 12 \ \texttt{D_Dstppipimu} & $\overline{B} \rightarrow D^{} (\rightarrow D^ \pi\pi) \mu \overline{\nu}\mu$ & $\textcolor{red}{n^{0}{D^{}}} \times N_{D \mu} \times \textcolor{red}{f_\text{guess}} \times f^0_{D^{*+}\pi\pi}$ & 13 \ \texttt{D_Dpipimu} & $\overline{B} \rightarrow D^{} (\rightarrow D^0 \pi\pi) \mu \overline{\nu}\mu$ & $\textcolor{red}{n^{0}{D^{}}} \times N_{D \mu} \times \textcolor{red}{f_\text{guess}} \times f^0_{D^{0}\pi\pi}$ & 14 \ \texttt{D_uD1mu} & $B^- \rightarrow D_1^0 \mu \overline{\nu}\mu$ & $\textcolor{blue}{\rho^\text{isospin}{D_1}} \times \textcolor{red}{n^{0}_{D^{}}} \times N_{D \mu} \times \textcolor{red}{f^{0}{D_1^0}} \times \mathcal{B}^{0}{D_1}$ & 15 \ \texttt{D_uD1mu_pipi} & $B^- \rightarrow D_1^0 (\rightarrow D^0 \pi\pi) \mu \overline{\nu}\mu$ & $\textcolor{blue}{\rho^\text{isospin}{D_1\pi\pi}} \times \textcolor{red}{n^{0}{D^{**}}} \times N{D \mu} \times \textcolor{red}{f^{0}{D_1^0}} \times \mathcal{B}^{0}{D_1\pi\pi}$ & 16 \ \texttt{D_uD2mu} & $B^- \rightarrow D_2^{0} \mu \overline{\nu}\mu$ & $\textcolor{blue}{\rho^\text{isospin}{D_2^}} \times \textcolor{red}{n^{0}{D^{**}}} \times N{D \mu} \times \textcolor{red}{f^{0}{D_2^{*0}}} \times \mathcal{B}^{0}{D_2^}$ & 17 \ \texttt{D_uD1pmu} & $B^- \rightarrow {D'1}^0 \mu \overline{\nu}\mu$ & $\textcolor{blue}{\rho^\text{isospin}{D'1}} \times \textcolor{red}{n^{0}{D^{**}}} \times N{D \mu} \times \textcolor{red}{f^{0}{D_1^{'0}}} \times \mathcal{B}^{0}{D'_1}$ & 18 \ \texttt{D_uD0mu} & $B^- \rightarrow {D^0}^0 \mu \overline{\nu}\mu$ & $\textcolor{blue}{\rho^\text{isospin}{D_1^*}} \times \textcolor{red}{n^{0}{D^{}}} \times N_{D \mu} \times \textcolor{red}{f^{0}{D_1^{*0}}} \times \mathcal{B}^{0}{D_1^}$ & 19 \ \texttt{D_sDs2mu} & $\overline{B}s \rightarrow D{s2}^ \mu \overline{\nu}\mu$ & $\textcolor{blue}{\mathcal{B}^{*}{D_{s2}^{*+}}} \times \textcolor{red}{\frac{f_s}{f_d}} \times \textcolor{red}{n^{0}_{D^{}}} \times \textcolor{red}{N^0_{s2}} \times N_{D \mu}$ & 20 \ \texttt{D_sDs1pmu} & $\overline{B}s \rightarrow D'{s1} \mu \overline{\nu}\mu$ & $\textcolor{blue}{\mathcal{B}^{*}{D_{s1}^{'+}}} \times \textcolor{red}{\frac{f_s}{f_d}} \times N_{D \mu} \times \textcolor{red}{n^{0}{D^{**}}} \times \textcolor{red}{N^0{s1'}}$ & 21 \ \texttt{D_dD1tau} & $\overline{B}^0 \rightarrow D_1 \tau \overline{\nu}\tau$ & $\textcolor{blue}{\rho^\text{isospin}{D_1}} \times \textcolor{blue}{\rho_{\mathcal{R}(D^{})}^0} \times \textcolor{red}{n^{0}_{D^{}}} \times N_{D \mu} \times \textcolor{red}{f^{0}{D_1}} \times \textcolor{red}{\epsilon{D_1}} \times \mathcal{B}^{0}{D_1} \times \textcolor{red}{\mathcal{R}(D^{**})\text{avg}^\text{raw}} \times \textcolor{red}{r({D_1})}$ & 22 \ \texttt{D_dD1tau_pipi} & $\overline{B}^0 \rightarrow D_1 (\rightarrow D^0 \pi\pi) \tau \overline{\nu}\tau$ & $\textcolor{blue}{\rho^\text{isospin}{D_1\pi\pi}} \times \textcolor{blue}{\rho_{\mathcal{R}(D^{})}^0} \times \textcolor{red}{n^{0}_{D^{}}} \times N_{D \mu} \times \textcolor{red}{f^{0}{D_1}} \times \textcolor{red}{\epsilon{D_1\pi\pi}} \times \mathcal{B}^{0}{D_1\pi\pi} \times \textcolor{red}{\mathcal{R}(D^{**})\text{avg}^\text{raw}} \times \textcolor{red}{r({D_1})}$ & 23 \ \texttt{D_dD2tau} & $\overline{B}^0 \rightarrow D^2 \tau \overline{\nu}\tau$ & $\textcolor{blue}{\rho^\text{isospin}_{D_2^}} \times \textcolor{blue}{\rho_{\mathcal{R}(D^{})}^0} \times \textcolor{red}{n^{0}_{D^{}}} \times N_{D \mu} \times \textcolor{red}{f^{0}{D_2^*}} \times \textcolor{red}{\epsilon{D_2^}} \times \mathcal{B}^{0}_{D_2^} \times \textcolor{red}{\mathcal{R}(D^{})\text{avg}^\text{raw}} \times \textcolor{red}{r({D_2^*})}$ & 24 \ \texttt{D_dD1ptau} & $\overline{B}^0 \rightarrow D'1 \tau \overline{\nu}\tau$ & $\textcolor{blue}{\rho^\text{isospin}{D'1}} \times \textcolor{blue}{\rho{\mathcal{R}(D^{})}^0} \times \textcolor{red}{n^{0}{D^{**}}} \times N{D \mu} \times \textcolor{red}{f^{0}{D'1}} \times \textcolor{red}{\epsilon{D'1}} \times \mathcal{B}^{0}{D'1} \times \textcolor{red}{\mathcal{R}(D^{**})\text{avg}^\text{raw}} \times \textcolor{red}{r({D'1})}$ & 25 \ \texttt{D_dD0tau} & $\overline{B}^0 \rightarrow D^*0 \tau \overline{\nu}\tau$ & $\textcolor{blue}{\rho^\text{isospin}{D_1^*}} \times \textcolor{blue}{\rho{\mathcal{R}(D^{})}^0} \times \textcolor{red}{n^{0}_{D^{}}} \times N_{D \mu} \times \textcolor{red}{f^{0}{D_1^*}} \times \textcolor{red}{\epsilon{D_1^}} \times \mathcal{B}^{0}_{D_1^} \times \textcolor{red}{\mathcal{R}(D^{})\text{avg}^\text{raw}} \times \textcolor{red}{r({D_1^*})}$ & 26 \ \texttt{D_uD1tau} & $B^- \rightarrow D_1^0 \tau \overline{\nu}\tau$ & $\textcolor{blue}{\rho^\text{isospin}{D_1}} \times \textcolor{blue}{\rho{\mathcal{R}(D^{})}^0} \times \textcolor{red}{n^{0}{D^{**}}} \times N{D \mu} \times \textcolor{red}{f^{0}{D_1^0}} \times \mathcal{B}^{0}{D_1} \times \textcolor{red}{\mathcal{R}(D^{})\text{avg}^\text{raw}} \times \textcolor{red}{r({D_1})}$ & 27 \ \texttt{D_uD1tau_pipi} & $B^- \rightarrow D_1^0 (\rightarrow D^0 \pi\pi) \mu \overline{\nu}\tau$ & $\textcolor{blue}{\rho^\text{isospin}{D_1\pi\pi}} \times \textcolor{blue}{\rho{\mathcal{R}(D^{})}^0} \times \textcolor{red}{n^{0}{D^{**}}} \times N{D \mu} \times \textcolor{red}{f^{0}{D_1^0}} \times \mathcal{B}^{0}{D_1\pi\pi} \times \textcolor{red}{\mathcal{R}(D^{})\text{avg}^\text{raw}} \times \textcolor{red}{r({D_1})}$ & 28 \ \texttt{D_uD2tau} & $B^- \rightarrow D_2^{*0} \tau \overline{\nu}\tau$ & $\textcolor{blue}{\rho^\text{isospin}{D_2^*}} \times \textcolor{blue}{\rho{\mathcal{R}(D^{})}^0} \times \textcolor{red}{n^{0}{D^{**}}} \times N{D \mu} \times \textcolor{red}{f^{0}{D_2^{*0}}} \times \mathcal{B}^{0}{D_2^} \times \textcolor{red}{\mathcal{R}(D^{**})_\text{avg}^\text{raw}} \times \textcolor{red}{r({D_2^})}$ & 29 \ \texttt{D_uD1ptau} & $B^- \rightarrow {D'1}^0 \tau \overline{\nu}\tau$ & $\textcolor{blue}{\rho^\text{isospin}{D'1}} \times \textcolor{blue}{\rho{\mathcal{R}(D^{**})}^0} \times \textcolor{red}{n^{0}{D^{}}} \times N_{D \mu} \times \textcolor{red}{f^{0}{D_1^{'0}}} \times \mathcal{B}^{0}{D'_1} \times \textcolor{red}{\mathcal{R}(D^{})\text{avg}^\text{raw}} \times \textcolor{red}{r({D'1})}$ & 30 \ \texttt{D_uD0tau} & $B^- \rightarrow {D^*0}^0 \tau \overline{\nu}\tau$ & $\textcolor{blue}{\rho^\text{isospin}{D_1^*}} \times \textcolor{blue}{\rho{\mathcal{R}(D^{})}^0} \times \textcolor{red}{n^{0}_{D^{}}} \times N_{D \mu} \times \textcolor{red}{f^{0}{D_1^{*0}}} \times \mathcal{B}^{0}{D_1^} \times \textcolor{red}{\mathcal{R}(D^{**})_\text{avg}^\text{raw}} \times \textcolor{red}{r({D_1^})}$ & 31 \ \texttt{D_dDDmu} & $\overline{B}^0 \rightarrow D^0 D_q (\rightarrow \mu \overline{\nu}\mu X') X$ & $\textcolor{magenta}{\rho^0{DDX,u/d}} \times f^0_{DDX} \times N_{D \mu} \times \textcolor{red}{f^0_{DDX,d}}$ & 32 \ \texttt{D_uDDmu} & $B^- \rightarrow D^0 D_q (\rightarrow \mu \overline{\nu}\mu X') X$ & $\textcolor{magenta}{\rho^0{DDX,u/d}} \times f^0_{DDX} \times N_{D \mu} \times \textcolor{red}{f^0_{DDX,u}}$ & 33 \ \texttt{D_dDDtau} & $\overline{B}^0 \rightarrow D^0 D_q (\rightarrow \tau \overline{\nu}\tau X') X$ & $\textcolor{blue}{\rho^0\text{isolation,\tau}} \times \textcolor{magenta}{\rho^0_{DDX,u/d}} \times \textcolor{red}{f^0_{DDX,\tau}} \times f^0_{DDX} \times N_{D \mu} \times \textcolor{red}{f^0_{DDX,d,\tau}}$ & 34 \ \texttt{D_uDDtau} & $B^- \rightarrow D^0 D_q (\rightarrow \tau \overline{\nu}\tau X') X$ & $\textcolor{blue}{\rho^0\text{isolation,\tau}} \times \textcolor{magenta}{\rho^0_{DDX,u/d}} \times \textcolor{red}{f^0_{DDX,\tau}} \times f^0_{DDX} \times N_{D \mu} \times \textcolor{red}{f^0_{DDX,u,\tau}}$ & 35 \ \texttt{D_comb} & $B^-$ comb. bkg. & $\textcolor{blue}{\rho^0_\text{$B$ comb}} \times \textcolor{red}{N^0_\text{$B$ comb}}$ & 36 \ \texttt{D_misID} & misID. & $\textcolor{blue}{\rho^0_\text{misID}} \times \textcolor{red}{N^0_\text{misID}}$ & 37 \ \bottomrule \end{tabular} }

::: ::: {.column width=50%}

\tightmargin\small

Total # of params: Ctrl: 117, Sig: 63
MC yields constrained relative to yields of normalization decays $$ \scriptsize \text{Yld MC} = \text{Yld norm} \times \text{Rel BF} \times \text{Rel sel eff} $$
- Example: $\Bm \rightarrow \Dz \taum\neutb$ in \Dz channel

\vspace{-15pt}\tiny $$ N_{D\mu} \times \left{ \RD \times \textcolor{red}{\underbrace{ \BRTauToMu \times \frac{ \epsilon{(\Bm \rightarrow \Dz \taum [\rightarrow \mun\neumb\neut] \neutb)} }{ \epsilon({\Bm \rightarrow \Dz \mun \neumb}) } }_{\equiv \fitRDEff} } \right} $$ \vspace{-15pt}

\small

Data-driven yields constrained (typically Gaussian) to expected yields derived from data ctrl samples $$ \scriptsize \text{Yld DDrv} = \text{Gaus constraint} \times \text{Expected yld} $$

::: :::

\begin{tikzpicture}[relative to page] \node[anchor=north, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, inner sep=3pt ] at (page cs:-0.5, 0.785) {\tiny Yield constraint scheme for \Dz ISO templates};

\node[anchor=north,
      draw=PepsiBlueLt,rounded corners,
      fill=PepsiBlueLt,fill opacity=.22,text opacity=1,
    ]
    at (page cs:-0.5, -0.88) {
        \tiny \bfseries floating; \textcolor{red}{fixed}; \textcolor{blue}{constrained}
    };

\end{tikzpicture}

Shape variations: quantify lack of knowledge

\vspace{1em} ::: columns ::: {.column width=25%} ::: ::: {.column width=25%} ::: ::: {.column width=25%} ::: ::: {.column width=25%} ::: :::

\vspace{1em} \visible<2->{ \begin{columns}[T] \begin{column}{0.33\textwidth} \includegraphics{./appendix/figs-supplemental-plots/pre-ctrl-fit/stacked/fit_result-stacked-D0-2os-q2.pdf} \end{column}

\begin{column}{0.33\textwidth} \includegraphics{./section/figs-fit-fit-results/ctrl-fit/stacked/fit_result-stacked-D0-2os-q2.pdf} \end{column}

\begin{column}{0.33\textwidth} \tightmargin\small

\begin{itemize} \tightlist \item Shape variations allow us to derive \textbf{phenomenological corrections}

\begin{itemize} \tightlist \item \Dz 2OS ctrl fit poor agreement to data w/o shape variations \item Fit quality mainly improved by (\Dstst_H) \qSq vars \end{itemize} \end{itemize} \end{column} \end{columns} }

\begin{tikzpicture}[relative to page] \node[anchor=south, execute at begin node=\setlength{\baselineskip}{7pt}, draw=normSig,rounded corners, fill=normSig,fill opacity=.5,text opacity=1, inner sep=2pt, text width=8em, align=center ] at (page cs:-0.72, 0.6) {\tiny \Dz (5), \Dstar (10), \Dstst (3+2) FF vars};

\node[anchor=south,
      execute at begin node=\setlength{\baselineskip}{7pt},
      draw=DststH,rounded corners,
      fill=DststH,fill opacity=.5,text opacity=1,
      inner sep=2pt, text width=8em, align=center
    ]
    at (page cs:-0.23, 0.6) {\tiny $\Dstst_H$ data-driven \qSq vars (3)};

\node[anchor=south,
      execute at begin node=\setlength{\baselineskip}{7pt},
      draw=DD,rounded corners,
      fill=DD,fill opacity=.5,text opacity=1,
      inner sep=2pt, text width=8em, align=center
    ]
    at (page cs:0.25, 0.6) {\tiny $DDX$ Dalitz-inspired data-driven vars (4)};

\node[anchor=south,
      execute at begin node=\setlength{\baselineskip}{7pt},
      draw=misID,rounded corners,
      fill=misID,fill opacity=.5,text opacity=1,
      inner sep=2pt, text width=8em, align=center
    ]
    at (page cs:0.75, 0.6) {\tiny misID decay-in-flight (1)};

% labels
\node[anchor=north west,
      draw=gray,rounded corners,
      fill=gray,fill opacity=.22,text opacity=1,
      inner sep=2pt,
    ]
    at (page cs:-0.91, 0.57) {\tiny \qSq};
\node[anchor=north west,
      draw=gray,rounded corners,
      fill=gray,fill opacity=.22,text opacity=1,
      inner sep=2pt,
    ]
    at (page cs:-0.42, 0.57) {\tiny \qSq};
\node[anchor=north west,
      draw=gray,rounded corners,
      fill=gray,fill opacity=.22,text opacity=1,
      inner sep=2pt,
    ]
    at (page cs:0.07, 0.57) {\tiny \qSq};
\node[anchor=north east,
      draw=gray,rounded corners,
      fill=gray,fill opacity=.22,text opacity=1,
      inner sep=2pt,
    ]
    at (page cs:0.93, 0.57) {\tiny \mmSq};

% Label for bot
\node<2->[anchor=north west,
      draw=gray,rounded corners,
      fill=gray,fill opacity=.22,text opacity=1,
      inner sep=2pt,
    ]
    at (page cs:-0.435, -0.11) {\tiny \qSq};
\node<2->[anchor=north west,
      draw=gray,rounded corners,
      fill=gray,fill opacity=.22,text opacity=1,
      inner sep=2pt,
    ]
    at (page cs:0.21, -0.11) {\tiny \qSq};

% Comment on included figs
\node[anchor=south east,
      execute at begin node=\setlength{\baselineskip}{7pt},
      draw=Dstst,rounded corners,
      fill=Dstst,fill opacity=.5,text opacity=1,
      inner sep=2pt, text width=5em, align=center
    ]
    at (page cs:-0.53, 0.115) {\tiny $B \rightarrow D_1$ FF var for BLR $\tau'$};

\node[anchor=south east,
      execute at begin node=\setlength{\baselineskip}{7pt},
      draw=DststH,rounded corners,
      fill=DststH,fill opacity=.5,text opacity=1,
      inner sep=2pt, text width=5em, align=center
    ]
    at (page cs:-0.045, 0.115) {\tiny $B \rightarrow \Dstst_H (\rightarrow \Dz)$};
\node[anchor=south east,
      execute at begin node=\setlength{\baselineskip}{7pt},
      draw=DD,rounded corners,
      fill=DD,fill opacity=.5,text opacity=1,
      inner sep=2pt, text width=5em, align=center
    ]
    at (page cs:0.445, 0.115) {\tiny $B \rightarrow \Dz D_q X$ lin/quad var};

% Fit
\node<2->[anchor=north,
      draw=gray,rounded corners,
      fill=gray,fill opacity=.22,text opacity=1,
      inner sep=2pt,
    ]
    at (page cs:-0.6, -0.82) {\tiny \Dz 2OS, no shape vars};
\node<2->[anchor=north,
      draw=gray,rounded corners,
      fill=gray,fill opacity=.22,text opacity=1,
      inner sep=2pt,
    ]
    at (page cs:0.03, -0.82) {\tiny \Dz 2OS, w/ shape vars};

\end{tikzpicture}

Fit result: \Dz ISO

::: columns ::: {.column width=33%} { width=90% } ::: ::: {.column width=33%} { width=90% } ::: ::: {.column width=33%} { width=90% } ::: :::

\vspace{0.4em}::: columns ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: :::

\vspace{0.3em}::: columns ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: :::

\begin{tikzpicture}[relative to page] \node[anchor=north west, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, inner sep=2pt ] at (page cs:-0.64, 0.72) {\tiny \mmSq}; \node[anchor=north west, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, inner sep=2pt ] at (page cs:-0.19, 0.72) {\tiny \el}; \node[anchor=north east, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, inner sep=2pt ] at (page cs:0.85, 0.72) {\tiny \qSq}; \end{tikzpicture}

Fit result: \Dstar ISO

::: columns ::: {.column width=33%} { width=90% } ::: ::: {.column width=33%} { width=90% } ::: ::: {.column width=33%} { width=90% } ::: :::

\vspace{0.4em}::: columns ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: :::

\vspace{0.3em}::: columns ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: :::

\begin{tikzpicture}[relative to page] \node[anchor=north west, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, inner sep=2pt ] at (page cs:-0.64, 0.72) {\tiny \mmSq}; \node[anchor=north west, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, inner sep=2pt ] at (page cs:-0.19, 0.72) {\tiny \el}; \node[anchor=north east, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, inner sep=2pt ] at (page cs:0.85, 0.72) {\tiny \qSq}; \end{tikzpicture}

Systematics (WIP)

\addcontentsline{toc}{chapter}{Systematics (WIP)}

::: columns ::: {.column width=50%}

Preliminary measurement of \RDX
- Introduction
- Event selection
- Trigger emulation for MC
- Data/MC corrections
- Fit
- Systematics (WIP)

::: ::: {.column width=50%}

Upgrade of the LHCb detector
- \color{gray}Overview of the LHCb upgrade
- The Upstream Tracker

::: :::

Current status of systematic studies

::: columns ::: {.column width=50%}

\vspace{0.5\baselineskip} \resizebox{\textwidth}{!}{ \begin{tabular}{r|c|c|c} \toprule {\bf Source} & {\bf $\sigma_{\RDst}$ [$\times 10^{-2}$]} & {\bf $\sigma_{\RD}$ [$\times 10^{-2}$]} & {\bf Correlation} \ \midrule %%%% \colorbox{yellow}{$B \rightarrow D^{()}\ellm\neulb$ form factors} & 0.53 (0.58) & 0.79 (2.37) & -0.71 (-0.80) \ %%%% \colorbox{yellow}{$B \rightarrow D^{**}\ellm\neulb$ form factors} & 0.25 (0.78) & 0.62 (1.01) & -0.85 (-0.10) \ %%%% Control sample shape parameters & 0.46 (0.87) & 0.96 (4.36) & TBD (-0.49) \ %%%% $\Lambda_b^0$ backgrounds & TBD (0.73) & TBD (1.16) & 0.00 (0.00) \ %%%% $DD$ model dependence\parnoteref{parnote:ctrl-shape-params} & TBD (0.63) & TBD (0.74) & 0.00 (0.00) \ %%%% \colorbox{yellow}{$B \rightarrow \Dstst\taum\neutb$ bkg} & 0.18 (0.17) & 0.49 (0.30) & -0.94 (0.78) \ %%%% \colorbox{yellow}{$B \rightarrow D^{()} \Dstst_s (\rightarrow \taum\neutb) X$ bkg} & 0.18 (0.25) & 0.62 (1.21) & -0.79 (0.59) \ %%%% \muon misID unfolding algorithm\parnoteref{parnote:ctrl-shape-params} & TBD (0.74) & TBD (1.19) & 1.00 (1.00) \ %%%% Coulomb correction to $\mathcal{R}(\Dstarp)$ vs. $\mathcal{R}(\Dstarz)$\parnoteref{parnote:ctrl-shape-params} & 0.27 (0.17) & 0.21 (0.3) & -1.00 (-1.00) \ %%%% \colorbox{yellow}{\muon misID decay-in-flight correction} & 0.14 (0.06) & 0.73 (0.16) & -0.95 (0.29) \ %%%% Combinatorial background shape\parnoteref{parnote:ctrl-shape-params} & TBD (0.03) & TBD (0.18) & 0.00 (0.00) \ %%%% Vertex resolution correction\parnoteref{parnote:ctrl-shape-params} & TBD (0.03) & TBD (0.21) & TBD (-0.25) \ %%%% Data/MC corrections (add.)\parnoteref{parnote:ctrl-shape-params} & TBD (0.40) & TBD (0.75) & 0.00 (0.00) \ %%%% \midrule Data/MC corrections (mul.)\parnote{ \label{parnote:mul} This is a multiplicative uncertainty. } & \makecell{$\text{TBD} \times \RDst$ \ ($1.16 \times \RDst$)} & \makecell{$\text{TBD} \times \RD$ \ ($0.91 \times \RD$)} & 0.00 (0.00) \ %%%% $\mathcal{B}(\taum \rightarrow \mun\neumb\neut)$ (ext. mea.)\parnoteref{parnote:mul} & \makecell{$0.23 \times \RDst$ \ ($0.23 \times \RDst$)} & \makecell{$0.23 \times \RD$ \ ($0.23 \times \RD$)} & 1.00 (1.00) \ \midrule %%%% Total systematics & TBD (2.03) & TBD (6.16) & TBD (-0.22) \ %%%% MC statistical & 0.91 (1.33) & 2.62 (4.00) & -0.72 (-0.30) \ %%%% Statistical & 1.94 (1.88) & 6.08 (6.01) & -0.07 (-0.53) \ \midrule Total (sys. + stats.) & TBD (3.07) & TBD (9.49) & TBD (-0.35) \ \bottomrule \end{tabular} }

::: ::: {.column width=50%}

\tightmargin\small

Most sys uncert included in fit as nuisances
- General idea: perform alternative fits w/ selected params fixed or different models
  - Subtract fitted uncertainties in quadrature $\rightarrow$ systematic uncertainty
Current status: have mechanism to eval most sys uncert
- Problem: fit not very stable, sometimes fixing params $\rightarrow$ different minimum
  - Data stat uncert larger likely because sys uncert underestimated ::: :::

\begin{tikzpicture}[relative to page] \node[anchor=north, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, ] at (page cs:-0.49, 0.73) { \tiny All uncert absolute, in $10^{-2}$; \colorbox{yellow}{uncert scale w/ data} }; \end{tikzpicture}

Outlook

\tightmargin ::: columns ::: {.column width=55%}

\vspace{1em} \small

Procedure to fit LHCb 2016 data developed
Fully portable to 2017 & 2018 (run 2)
- Perform data/MC correction separately for each year
  - Each year has different trigger thresholds
- Weight corrected templates by lumi & fit combined 2016--2018 data
All key elements of \RDX measurement in run 2 data developed
- Some work on systematic uncertainties remaining
- Challenge to reduce systematics to level of small stat uncert

::: ::: {.column width=45%}

\centering { width=80% }

\small

Stat & sys uncert comparable, some sys scales w/ data
- Can use more data!

\vspace{2.5em}

::: :::

\begin{tikzpicture}[relative to page] \node[anchor=north] at (page cs:0.53, -0.60) { \tiny \begin{tabular}{r|c|c} & run 1 & run 2 target \ \midrule \RD & $\pm 14%\text{stat} \pm 15%\text{sys}$ & $\pm 7%\text{stat} \pm 8%\text{sys}$ \ \RDst & $\pm 6.4%\text{stat} \pm 8.2%\text{sys}$ & $\pm 3%\text{stat} \pm 4%\text{sys}$ \ \end{tabular} }; \end{tikzpicture}

Overview of the LHCb upgrade

::: columns ::: {.column width=50%}

Preliminary measurement of \RDX
- Introduction
- Event selection
- Trigger emulation for MC
- Data/MC corrections
- Fit
- Systematics (WIP)

::: ::: {.column width=50%}

Upgrade of the LHCb detector
- Overview of the LHCb upgrade
- \color{gray} The Upstream Tracker

::: :::

LHCb readout bottleneck in run 1 & 2

::: columns ::: {.column width=50%}

\tightmargin\small\vspace{0.5em}

LHCb doesn't utilize LHC's peak luminosity
- CMS: $1.5 \times 10^{34}$ \lumiInsta
- LHCb: $4 \times 10^{32}$ \lumiInsta, ~1/40 of CMS
  - Lumi levelling by de-focusing beams
Main bottleneck: detector readout rate at 1 MHz
- LHC bunch-crossing rate: 40 MHz, collision rate: ~30 MHz
Hardware triggers (cut on \pt, $E_T$) to keep readout rate constant
- Higher lumi $\rightarrow$ harder cuts
- Hadronic triggers saturate, no benefit from increase in lumi
- Run 2 int. lumi: ~2 \ifb/year

::: ::: {.column width=50%}

\centering { width=70% }

\begin{itemize} \tightlist\small \item<2-> \textbf{Solution: readout at 40 MHz so hardware triggers can be removed}

\begin{itemize} \tightlist \item \textbf{LHCb run 3 lumi}: (2 \times 10^{33}) \lumiInsta \item \textbf{First (hardware) triggers-less detector in a hadron collider} \end{itemize} \end{itemize}

::: :::

The LHCb detector in run 3

::: columns ::: {.column width=60%}

\begin{onlyenv}<1-2> \vspace{2em} \includegraphics<1>[width=\textwidth]{./chapter/figs-detector/lhcb_detector_view.pdf} \includegraphics<2>[width=\textwidth]{./chapter/figs-lhcb-upgrade-overview/lhcb_detector_view_run3.pdf}

\begin{tikzpicture}[relative to page] \node<1>[anchor=north, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, ] at (page cs:-0.5, 0.65) {\small run 1--2\vphantom{3}};

\node<2>[anchor=north,
      draw=PepsiBlueLt,rounded corners,
      fill=PepsiBlueLt,fill opacity=.22,text opacity=1,
    ]
    at (page cs:-0.5, 0.65) {\small run 3\vphantom{12}};

\end{tikzpicture} \end{onlyenv}

::: ::: {.column width=40%}

\tightmargin\small

Full detector readout rate upgraded to 40 MHz
New tracking system: better resolution & more radiation tolerant
- VELO $\rightarrow$ Pixelated VELO
- TT $\rightarrow$ UT
  - Major contrib. from Maryland
- T-stations $\rightarrow$ SciFi
Reduce RICH occupancy: maintain good PID w/ higher lumi
Remove M1, PS & SPD (used for HW trigger): less material before ECAL $\rightarrow$ better $E$ resolution

::: :::

Concept arts of the upgraded tracking system

\centering \begin{figure}[H] \centering \begin{subfigure}[t]{0.6\textwidth} \centering \includegraphics[width=\textwidth]{./chapter/figs-lhcb-upgrade-overview/tracking/velo_upgrade.pdf} \end{subfigure}

%%%%
\begin{subfigure}[t]{0.35\textwidth}
    \centering
    \includegraphics[width=\textwidth]{./chapter/figs-lhcb-upgrade-overview/tracking/ut_upgrade.pdf}
\end{subfigure}
\hspace{3em}
\begin{subfigure}[t]{0.35\textwidth}
    \centering
    \includegraphics[width=\textwidth]{./chapter/figs-lhcb-upgrade-overview/tracking/scifi_upgrade.pdf}
\end{subfigure}

\end{figure}

\begin{tikzpicture}[relative to page] \node[anchor=west, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, ] at (page cs:0.58, 0.42) {\small Pixelated VELO};

\node[anchor=north east,
      draw=PepsiBlueLt,rounded corners,
      fill=PepsiBlueLt,fill opacity=.22,text opacity=1,
    ]
    at (page cs:-0.75, -0.5) {\small UT};

\node[anchor=north west,
      draw=PepsiBlueLt,rounded corners,
      fill=PepsiBlueLt,fill opacity=.22,text opacity=1,
    ]
    at (page cs:0.75, -0.5) {\small SciFi};

\end{tikzpicture}

The upgraded trigger scheme

\tightmargin ::: columns ::: {.column width=50%}

{ height=12em } { height=12em }

::: ::: {.column width=50%}

\small

Upgraded LHCb removes HW trigger
High-level SW triggers: HLT1 & HLT2
HLT1 w/ event builder (resp. for collecting data from detector)
- Each event builder hosts 2 GPUs for HLT1
  - Ideally suited for track reco
  - 30x reduction in inter-communication bandwidth

::: :::

The Upstream Tracker

::: columns ::: {.column width=50%}

Preliminary measurement of \RDX
- Introduction
- Event selection
- Trigger emulation for MC
- Data/MC corrections
- Fit
- Systematics (WIP)

::: ::: {.column width=50%}

Upgrade of the LHCb detector
- Overview of the LHCb upgrade
- The Upstream Tracker

::: :::

Overview of the Upstream Tracker

::: columns ::: {.column width=60%}

{ width=15% } { width=32% } \hspace{0.5em} { width=45% }

::: ::: {.column width=40%}

\tightmargin\small

UT: a silicon-strip detector
- 4 detection layers, $x$-$u$-$v$-$x$ configuration
- Better resolution near beam pipe
- Closer to beam pipe (circular cutout)
Detector box
- Staves
  - Sensors & SALT readout ASIC
PEPI
- Backplane (BP)
- DCB
Service Bay
- LVR

::: :::

\begin{tikzpicture}[relative to page] \node[anchor=north west,inner sep=0pt] at (page cs:0.78, 0.75) { \includegraphics[width=0.11\textwidth]{./slides-figures/xuvx_schematic.pdf} };

\node[anchor=north west,inner sep=0pt]
    at (page cs:0.54, -0.426) {
        \includegraphics[width=0.23\textwidth]{./chapter/figs-ut-upgrade/stave/stave_hybrid_closeup.jpg}
    };

\node[anchor=north,
      draw=Green,rounded corners,inner sep=2pt,
      fill=Green,fill opacity=.22,text opacity=1,text=white]
    at (page cs:-0.49,-0.6) {\footnotesize stave};

\node[anchor=north,
      draw=Green,rounded corners,inner sep=2pt,
      fill=Green,fill opacity=.22,text opacity=1,text=white]
    at (page cs:-0.02,-0.6) {\footnotesize sensor};

\node[anchor=north,
      draw=Green,rounded corners,inner sep=2pt,
      fill=Green,fill opacity=.22,text opacity=1,text=white]
    at (page cs:0.66,-0.6) {\footnotesize SALT};

\end{tikzpicture}

Backplane & LVR

\tightmargin ::: columns ::: {.column width=50%}

\small

Backplane (BP), 30
- Deliver power & data between connectors
- No active component (i.e IC)
- 28 layers of PCB, at the limit of manufacturability
  - Typical PC motherboard: 6--8 layers of PCB

::: ::: {.column width=50%}

Low Voltage Regulator (LVR), ~240
- Supply power to DCB & SALT ASIC (~10m away)
- Remote sensing: ensure device-side voltage is kept at specified voltage
  - DCB: 1.5 V, DCB opt: 2.5 V, SALT: 1.25 V

::: :::

\vspace{8.5em} \begin{tikzpicture}[relative to page] \node[anchor=north,inner sep=0pt] at (page cs:-0.74,-0.1) { \includegraphics[height=9em]{./chapter/figs-ut-upgrade/backplane/backplane_trace.pdf} }; \node[anchor=north,inner sep=0pt] at (page cs:-0.26,-0.1) { \includegraphics[height=9em]{./chapter/figs-ut-upgrade/backplane/backplane_compressed.jpg} };

\node[anchor=north,inner sep=0pt] at (page cs:0.26,-0.1) {
    \includegraphics[height=9em]{./chapter/figs-ut-upgrade/lvr/lvr_top_view.pdf}
};
\node[anchor=north,inner sep=0pt] at (page cs:0.74,-0.1) {
    \includegraphics[height=9em]{./chapter/figs-ut-upgrade/lvr/lvr_bot_view.pdf}
};

\end{tikzpicture}

DCB

\tightmargin ::: columns ::: {.column width=50%}

\small

Data Control Board (DCB), ~250
- Clock distribution to SALT
- Control command distribution to SALT
- Serialization of SALT readout
- Transmission of serialized data
- Telemetry (thermistor readouts)

::: ::: {.column width=50%}

\small

1 master GBTx (clock & ctrl dist)
- 1 VTRx (bi-dir opt comm to ctrl sys)
1 GBT SCA (ctrl dist & ADC readout)
6 data GBTx (data serialization)
- 3 VTTx (uni-dir opt transmission of data)

::: :::

\vspace{1.5em} ::: columns ::: {.column width=50%} { width=90% } ::: ::: {.column width=50%} { width=85% } ::: :::

\begin{tikzpicture}[relative to page] \node[anchor=north west, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, ] at (page cs:-0.92, -0.72) { \footnotesize Later: DCB functionality validation & QA }; \end{tikzpicture}

DCB functionality validation & QA

::: columns ::: {.column width=45%}

\tightmargin\small

DCB: error-free for $\mathcal{O}(10^{15})$ bits
- Configure data GBTxs to generate & transmit PRBS
- Use MiniDAQ to check error at firmware level
- Produce eye-diagram overnight: wide-open "eye" $\rightarrow$ low-jitter

\centering { width=85% }

::: ::: {.column width=55%}

\tightmargin\small

QA at Maryland: reverse-engineered a cli program to init & ctrl DCB in batch
- dcbutil.py write 1c 1 -g 3
- 270 produced & tested at Maryland
QA at CERN: wrote a one-click panel for test
- 260 shipped to CERN
- QA'ed them all, ensure no damage in shipping

{ width=40% } { width=40% }

::: :::

\begin{tikzpicture}[relative to page] \node[anchor=north west, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, ] at (page cs:0.7, 0.525) {\tiny\texttt{write I2C reg}}; \end{tikzpicture}

UMD members working on UT

::: columns ::: {.column width=50%} \centering { height=8em } ::: ::: {.column width=50%} \centering { height=8em } ::: :::

\vspace{2em} ::: columns ::: {.column width=50%} \centering { height=8em } ::: ::: {.column width=50%} \centering { height=8em } ::: :::

Conclusion

Implemented main framework for \RDX run 2 analysis
- Created infrastructure to handle large simulated sample
- Developed data/MC corrections
- Fitted 2016 data
- Have the capability of evaluating most systematics
- Next steps
  - Further study of systematics & fit convergence
  - Expand to 2017--2018 data
Major contribution to LHCb UT upgrade
- DCB development & QA
- Installation effort at CERN

Backup

The CKM matrix

{ width=60% }

\tightmargin\small \small

$\lambda \approx 0.04$ small compared to 1 \small
$B$ meson has an unusually long life time

\Dstst and $\Dstst_H$ cascade decays

\centering

MisID background

\begin{equation*} \begin{pmatrix*}[l] \tilde{n}{\hat{\pi}} \ \tilde{n}{\hat{K}} \ \tilde{n}{\hat{p}} \ \tilde{n}{\hat{e}} \ \tilde{n}{\hat{g}} \ \end{pmatrix*} = \begin{pmatrix*}[l] \misEff[\pi]{\hat{\pi}} & \misEff[K]{\hat{\pi}} & \misEff[p]{\hat{\pi}} & \misEff[e]{\hat{\pi}} & \misEff[g]{\hat{\pi}} \ \misEff[\pi]{\hat{K}} & \misEff[K]{\hat{K}} & \misEff[p]{\hat{K}} & \misEff[e]{\hat{K}} & \misEff[g]{\hat{K}} \ \misEff[\pi]{\hat{p}} & \misEff[K]{\hat{p}} & \misEff[p]{\hat{p}} & \misEff[e]{\hat{p}} & \misEff[g]{\hat{p}} \ \misEff[\pi]{\hat{e}} & \misEff[K]{\hat{e}} & \misEff[p]{\hat{e}} & \misEff[e]{\hat{e}} & \misEff[g]{\hat{e}} \ \misEff[\pi]{\hat{g}} & \misEff[K]{\hat{g}} & \misEff[p]{\hat{g}} & \misEff[e]{\hat{g}} & \misEff[g]{\hat{g}} \ \end{pmatrix*} \begin{pmatrix*}[l] \tilde{n}{{\pi}} \ \tilde{n}{{K}} \ \tilde{n}{{p}} \ \tilde{n}{{e}} \ \tilde{n}{{g}} \ \end{pmatrix*} \end{equation*}

Know tagged yields (left) and true $\rightarrow$ tag eff from dedicated data samples
Can find true yields w/ unfolding
- Can't use matrix inversion b.c. sensitive to statistical fluctuations

Relevant form factor parameterizations

\tightmargin

::: columns ::: {.column width=50%}

\small

ISGW2
- Fully predictive
  - No free parameter
- \textbf{Doesn't describe data well}
CLN
- Based on BGL, \textbf{apply HQET to reduce num of params}
  - \Dz: 3 params
  - \Dstar: 5 params
- Some parameters too closely cross-constrained

::: ::: {.column width=50%}

\small

BGL
- Based on dispersion relations
  - \Dz: 5 params
  - \Dstar: 10 params
- Analytically continue FFs as complex functions $\rightarrow$ expandable
- \textbf{Model independent} until truncate series
- \textbf{Many free parameters} restricted from lattice QCD + data
BLR
- Apply HQET to \Dstst
  - $D^{1/2+} (D^*_0, D'_1)$: 3 params, 1 overall norm
  - $D^{3/2+} (D_1, D^*_2)$: 4 params, 1 overall norm
- \textbf{Offer parameters fitted from data}

::: :::

\begin{tikzpicture}[relative to page] \node[anchor=north west, execute at begin node=\setlength{\baselineskip}{7pt}, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, text width=15em,align=center ] at (page cs:-0.95, -0.53) { \footnotesize Only ISGW2 & CLN implemented in MC simulation program \texttt{EvtGen} };

\node[anchor=north west,
      execute at begin node=\setlength{\baselineskip}{7pt},
      draw=PepsiBlueLt,rounded corners,
      fill=PepsiBlueLt,fill opacity=.22,text opacity=1,
    ]
    at (page cs:0.67, -0.38) {
        \tiny approximately
    };

\end{tikzpicture}

Fit result: $D^0$ 1OS

\vspace{-0.5em}::: columns ::: {.column width=33%} { width=90% } ::: ::: {.column width=33%} { width=90% } ::: ::: {.column width=33%} { width=90% } ::: :::

\vspace{0.4em}::: columns ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: :::

\vspace{0.3em}::: columns ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: :::

Fit result: $D^0$ 2OS

\vspace{-0.5em}::: columns ::: {.column width=33%} { width=90% } ::: ::: {.column width=33%} { width=90% } ::: ::: {.column width=33%} { width=90% } ::: :::

\vspace{0.4em}::: columns ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: :::

\vspace{0.3em}::: columns ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: :::

Fit result: $D^0$ DD

\vspace{-0.5em}::: columns ::: {.column width=33%} { width=90% } ::: ::: {.column width=33%} { width=90% } ::: ::: {.column width=33%} { width=90% } ::: :::

\vspace{0.4em}::: columns ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: :::

\vspace{0.3em}::: columns ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: :::

Fit result: $D^*$ 1OS

\vspace{-0.5em}::: columns ::: {.column width=33%} { width=90% } ::: ::: {.column width=33%} { width=90% } ::: ::: {.column width=33%} { width=90% } ::: :::

\vspace{0.4em}::: columns ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: :::

\vspace{0.3em}::: columns ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: :::

Fit result: $D^*$ 2OS

\vspace{-0.5em}::: columns ::: {.column width=33%} { width=90% } ::: ::: {.column width=33%} { width=90% } ::: ::: {.column width=33%} { width=90% } ::: :::

\vspace{0.4em}::: columns ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: :::

\vspace{0.3em}::: columns ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: :::

Fit result: $D^*$ DD

\vspace{-0.5em}::: columns ::: {.column width=33%} { width=90% } ::: ::: {.column width=33%} { width=90% } ::: ::: {.column width=33%} { width=90% } ::: :::

\vspace{0.4em}::: columns ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: :::

\vspace{0.3em}::: columns ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: ::: {.column width=25%} { width=90% } ::: :::

Fitted yields

\vspace{0.5em} \begin{table}[H] \centering

\begin{subtable}[b]{0.5\textwidth} \centering

\begin{tabular}[b]{lr} \hline \bfseries Group & \bfseries Yields \ \hline norm. ($D^0\mu$) & 445,095 \ norm. ($D^{*+}\mu$) & 115,124 \ norm. ($D^{*0}\mu$) & 1,176,363 \ sig. & \colorbox{black}{XXX} \ $D^{}$ & 167,968 \ $D^{}$ heavy & 38,147 \ $D_s$ & 5,285 \ $DD$ & 96,236 \ comb. bkg. & 20,596 \ misID & 56,596 \ \hline \end{tabular} \end{subtable}% %%%% \begin{subtable}[b]{0.5\textwidth} \centering

\begin{tabular}[b]{lr} \hline \bfseries Group & \bfseries Yields \ \hline norm. ($D^{*+}\mu$) & 427,783 \ sig. & \colorbox{black}{XXX} \ $D^{}$ & 37,181 \ $D^{}$ heavy & 12,804 \ $D_s$ & 3,730 \ $DD$ & 33,588 \ comb. bkg. & 16,906 \ misID & 8,175 \ \hline \end{tabular} \end{subtable}

\end{table}

\begin{tikzpicture}[relative to page] \node[anchor=north, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, ] at (page cs:-0.45, -0.74) {\scriptsize \Dz channel}; \node[anchor=north, draw=PepsiBlueLt,rounded corners, fill=PepsiBlueLt,fill opacity=.22,text opacity=1, ] at (page cs:0.45, -0.74) {\scriptsize \Dstar channel};

\node[anchor=north,
      draw=PepsiBlueLt,rounded corners,
      fill=PepsiBlueLt,fill opacity=.22,text opacity=1,
    ]
    at (page cs:-0.45, 0.74) {\scriptsize Analysis blinded};

\end{tikzpicture}

Files

slides.md

Latest commit

History

slides.md

File metadata and controls

Outline

Introduction

The standard model and beyond

Precision measurements & tests of LFU

Testing LFU with \RDX

About this analysis

The Large Hadron Collider (LHC)

The LHCb detector

Event selection

Signal and normalization {.fragile}

Selection of \Dz\muon and \Dstarp\muon

\Dz\muon (\Dz channel)

\Dstarp\muon (\Dstar channel)

\UBDT

Feed down from \Dstarp\muon and \Dstarz\muon

\Dstarp\muon: $\Bzb \rightarrow \Dstarp (\rightarrow \Dz\pip) l^- \neulb$

\Dstarz\muon: $\Bm \rightarrow \Dstarz (\rightarrow \Dz\piz) l^- \neulb$

Background contributions

The isolation BDT

Signal and control skims (sub-samples)

Key kinematic variables: \mmSq, \el, \qSq

Rest frame approximation

Trigger emulation for MC

Tracker-only MC

Emulate L0

Emulate HLT

Data/MC corrections

Procedure overview

Form factor reweighting

Final reweighting

Fit

Overall fit strategy

Templates in the fit

Fit model

Shape variations: quantify lack of knowledge

Fit result: \Dz ISO

Fit result: \Dstar ISO

Systematics (WIP)

Current status of systematic studies

Outlook

Overview of the LHCb upgrade

LHCb readout bottleneck in run 1 & 2

The LHCb detector in run 3

Concept arts of the upgraded tracking system

The upgraded trigger scheme

The Upstream Tracker

Overview of the Upstream Tracker

Backplane & LVR

DCB

DCB functionality validation & QA

UMD members working on UT

Conclusion

Backup

The CKM matrix

\Dstst and $\Dstst_H$ cascade decays

MisID background

Relevant form factor parameterizations

Fit result: $D^0$ 1OS

Fit result: $D^0$ 2OS

Fit result: $D^0$ DD

Fit result: $D^*$ 1OS

Fit result: $D^*$ 2OS

Fit result: $D^*$ DD

Fitted yields