[Paper Review] Implementation and application of kinematic vertex fitting in the software environment of ILD

Fabian Moser, W. Waltenberger+2 more|ArXiv.org|Jan 26, 2009

Simulation and Modeling Applications3 citations

TL;DR

This paper presents the integration of kinematic vertex fitting into the ILD software framework via the MarlinRave plug-in, enabling constrained reconstruction of $WW$ and $ZZ$ decays using RAVE's vertexing toolkit. By applying energy-momentum conservation and a similar-mass hypothesis, the method improves $W$ mass resolution, achieving a reconstructed mean of $79.58 \pm 0.27$ GeV with a precision of 0.014 GeV in 4-year ILC operation.

ABSTRACT

The vertex reconstruction toolkit RAVE has been extended by an option for the inclusion of kinematic constraints, and embedded into the ILD analysis framework Marlin. The new tools have been tested with an exemplary reconstruction of WW and ZZ decays. The presented results show the improvements achieved in precision of the fitted masses, and demonstrate the usage and functionality of the toolkit.

Motivation & Objective

• To enhance precision in reconstructing $W^+W^-$ and $ZZ$ decay vertices in the ILC's ILD detector by incorporating kinematic constraints.
• To develop a reusable, experiment-independent vertex fitting toolkit compatible with the ILD analysis framework Marlin.
• To reduce jet association ambiguity in 4-jet final states by applying kinematic constraints and a statistical selection criterion based on mass similarity.
• To demonstrate improved mass resolution and reduced systematic errors in $W$ and $Z$ boson reconstruction using simulated ILC data.

Proposed method

• Extended the RAVE library with kinematic vertex fitting algorithms originally from CMSSW, enabling constrained geometric reconstruction.
• Developed the MarlinRave plug-in to integrate RAVE’s capabilities into the ILD analysis framework, exposing two processors: RaveVertexing and RaveKinematics.
• Applied energy and 3-momentum conservation constraints to four-jet final states via equations (3) and (4), enforcing $\sum E_i = \sqrt{s}$ and $\sum \vec{p}_i = 0$.
• Implemented a 'similar-mass hypothesis' using a likelihood function that models two $W$ masses as drawn from a common distribution with mean $\bar{m}$ and scale $\sigma_t \approx 9$ GeV.
• Formulated a combined objective function (equation 6) minimizing the sum of the kinematic fit $\chi^2$ and mass similarity penalty, used to select the optimal jet pairing.
• Used the pseudo-\chi^2 probability from the objective function as a selection criterion to rank and select the best jet combination among three possible pairings.

Experimental results

Research questions

• RQ1 Can kinematic constraints significantly improve the precision of $W$ and $Z$ boson mass reconstruction in $e^+e^-$ collisions at the ILC?
• RQ2 How effective is the similar-mass hypothesis in reducing jet pairing ambiguity in 4-jet final states compared to standard kinematic fitting?
• RQ3 To what extent do energy-momentum conservation and mass similarity constraints reduce systematic errors in vertex reconstruction for $WW$ and $ZZ$ decays?
• RQ4 What is the achievable mass resolution for $W$ bosons after applying constrained fitting in a realistic ILC detector simulation?

Key findings

• The kinematic vertex fitting with energy-momentum conservation and similar-mass constraint reduced type I and type II jet association errors to 1.6% and 0.9%, respectively.
• After applying a 130 GeV mass cut and the similar-mass constraint, 979 events remained from an initial sample of 1008 $WW$ events, indicating high selection efficiency.
• The reconstructed $W$ mass distribution had a mean of $79.58 \pm 0.27$ GeV and a width of $11.4 \pm 0.2$ GeV, closely matching the true $W$ mass of 80.32 GeV.
• Projected to 4 years of ILC operation (664k events), the method achieves a $W$ mass resolution of 0.014 GeV, demonstrating high-precision potential.
• For $ZZ$ decays, the inclusion of kinematic constraints significantly improved mass resolution compared to unconstrained reconstruction, despite initial state radiation not being modeled.
• The method successfully accounts for redundant degrees of freedom by combining four kinematic constraints with a statistical mass similarity prior, yielding a robust and stable fit.