W Mass Reconstruction

We have generated W bosons isotropically in the momentum range

$\begin{displaymath} \begin{array} {lll} \nonumber 0 \le P_W \le 500 {\rm GeV}\end{array}\end{displaymath}$

with $\Gamma_W=0$ .The generated events are passed through the JLC-standard-detector simulation and the invariant mass of the final state particles is calculated for each event without any selection cuts. Fig.3.1 compares the invariant mass distribution for calorimetric information only and that after the new cluster-track matching.

**Figure 3.1:** Reconstructed W mass distributions for (left) calorimetric information only and (right) combined track information. In the cluster-track matching 2- $\sigma$ cut is used to delete global clusters.
$\begin{figure} \centerline{ \epsfxsize=5.5cm \epsfbox{mw_cal.epsf} \epsfxsize=5.5cm \epsfbox{mw_cmb.epsf}}\end{figure}$

Fig.3.2 corresponds to blown-up of Fig.3.1, again (left) for calorimetric information only and (right) for combined track information. The figure tells us that, after the new cluster-track matching we have

$\begin{displaymath} \begin{array} {lll} \nonumber < \Delta M_W \gt \simeq -1.5 {\rm GeV},\end{array}\end{displaymath}$

not too good showing the 2- $\sigma$ cut is to much, while the mass resolution improves to

$\begin{displaymath} \begin{array} {lll} \nonumber \sigma_W \simeq 1.8 {\rm GeV},\end{array}\end{displaymath}$

which is quite good compared to the mass resolution in excess of $4 {\rm GeV}$ with the calorimetric information only.

**Figure 3.2:** Blown-up figure of Fig.3.1, where the smooth curves are the result of Gaussian-fits: (left) calorimetric information only and (right) combined track information.
$\begin{figure} \centerline{ \epsfxsize=5.5cm \epsfbox{mwblow_cal.epsf} \epsfxsize=5.5cm \epsfbox{mwblow_cmb.epsf}}\end{figure}$

When we examine the mass distribution as a function of momentum, we get Fig.3.3.

**Figure 3.3:** Reconstructed W mass plotted against W momentum and its standard deviation and center value as functions of the W momentum: (left) calorimetric information only and (right) combined track information.
$\begin{figure} \centerline{ \epsfxsize=5.5cm \epsfbox{mwp-a_cal.epsf} \epsfxsiz... ...epsfbox{mwp-c_cal.epsf} \epsfxsize=5.5cm \epsfbox{mwp-c_cmb.epsf} }\end{figure}$

We can see that both the resolution and the center value are momentum-dependent even after the new cluster-track matching: $\sigma_W \simeq 1.6 {\rm GeV}$ at best, while the center value decreases with momentum, showing that the rate of cluster misdeletions goes up with momentum. The mass resolution and the systematic shift of the center value depend on the choice of the $n \cdot \sigma$ cut.