# Improving the spatial resolution of NICA/MPD ECAL with new reconstruction methods

## TL;DR

This paper improves the spatial resolution of the Tsinghua ECal modules using new reconstruction methods, including deep learning, achieving a 30% enhancement over traditional techniques in beam test conditions.

## Contribution

Introduces a deep learning-based reconstruction method that significantly enhances the position resolution of the NICA/MPD ECAL modules.

## Key findings

- Deep learning method reduces position resolution to below 3.8 mm for 1.6 GeV electrons.
- Achieves approximately 30% improvement over traditional charge center of gravity method.
- Validates the new reconstruction approach with beam test data from DESY.

## Abstract

A Shashlyk-type electromagnetic calorimeter (ECal) will be used in the Multi-purpose Detector at Nuclotron-based Ion Collider facility to study the properties of nuclear matter. In this experiment, the ECal detector is responsible for measuring the energy and position of the incident particles, and identifying them with the information obtained from itself and other detectors. This paper analyzes the position resolution of the Tsinghua ECal modules using the data from a beam test in DESY. Several reconstruction methods are studied in detail. With a deep learning based algorithm, the position resolution of the prototypes achieves less than 3.8 mm for 1.6 GeV electron beam, which is improved by about 30% compared to that of traditional charge center of gravity method.