Abstract:Identification of particles generated by ion collisions in the NICA collider is one of the basic functions of the Multipurpose Detector (MPD). The main means of identification in MPD are the time-of-flight system (TOF) and the time-projection chamber (TPC). The article considers the optimization of the algorithms of particles identification by these systems. Under certain conditions, the use of the statistical Bayesian approach has made it possible to achieve an optimal ratio of the efficiency of particle identification and contamination by incorrectly defined particles.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to optimize the particle identification algorithm in the Multi - Purpose Detector (MPD) of the NICA collider, especially in the identification of low - momentum particles, and improve the efficiency and purity of particle identification. It is mentioned in the paper that although the traditional "n - sigma" method can achieve effective separation of particles, the identification purity of some types of particles (such as K - mesons) in the high - momentum region is not high enough. Therefore, the author introduced the Bayesian statistical method to optimize the particle identification process. Specifically, by combining the data of the Time - of - Flight system (TOF) and the Time Projection Chamber (TPC), the paper re - evaluates the probability threshold of particle identification using the Bayesian method, thereby reducing the number of misidentified particles while ensuring the identification efficiency. This method is particularly suitable for research that requires high - purity separation of specific types of particles, such as the study of specific particle decay channels. The paper also describes an iterative method to find the prior probabilities (priors), which are used to adjust the parameters in the Bayesian identification algorithm. Through multiple iterations, the prior probabilities can converge to a stable value, making the data processing results more independent of the initially selected prior probabilities. This helps to improve the quality of particle identification, especially when considering the characteristics of real detectors and background particles.

Bayesian Approach to Particles Identification in the MPD Experimen

Towards a Realistic Monte Carlo Simulation of the MPD Detector at NICA

Improving Photoelectron Counting and Particle Identification in Scintillation Detectors with Bayesian Techniques

Reconstruction of moments of particle distributions with Identity Method at MPD

Track reconstruction and particle identification developments for a study of event-by-event fluctuations in heavy ion collisions at NICA

Prospects of the MPD Detector for Measuring the Spectra of Identified Hadrons in Collisions at Energy GeV

Performance studies of anisotropic flow with MPD at NICA

Status of the MPD Experiment at JINR

Trends in particle and nuclei identification techniques in nuclear physics experiments

Status of the NICA/MPD project

Charged particle identification with the liquid xenon calorimeter of the CMD-3 detector

Performance study for anisotropic flow measurements in the MPD (NICA) experiment with fixed target

Machine-learning-based particle identification with missing data

GRB990123: multiwavelength afterglow study

Particle-Set Identification method to study multiplicity fluctuations

Performance of the MPD experiment for the anisotropic flow measurement

Particle Identification Using Artificial Neural Networks at Besiii

Particle Identification with the ePIC detector at the EIC

TPC tracking and particle identification in high-density environment

The comparison of methods for anisotropic flow measurements with the MPD Experiment at NICA