Data analysis with deep learning technique and real-time event reconstruction in MUonE

Student: Katherine Ferraby
Supervisors: Thomas Teubner, Themis Bowcock

A new project, MUonE, has been proposed at CERN to measure the shape of the differential cross section of μ-e elastic scattering as a function of the spacelike squared momentum transfer. The proposed detector is a repetition of 40 identical modules (called stations), each consisting of a 1.5 cm thick layer of Be (or C) coupled to 3 Si tracking planes, located at a relative distance of one meter from each other with intermediate air gaps.

A downstream particle identifier made by a calorimeter for the electrons and a muon system for the muons (a filter plus active planes) is equired to solve the muon-electron ambiguity for electron scattering angles around (2–3) mrad and background suppression. Silicon (Si) strip sensors being produced for the CMS Tracker upgrade (in the so-called 2S configuration) are considered as baseline choice for the tracking planes. The angles of outgoing electrons and muons are measured with high accuracy by Si detectors. The beam intensity provides the required event yield. With an average rate of 1.3´107 μ sec−1, 4´1012 elastic events will be produced in about 3 years of data taking and allow to reach a statistical uncertainty on aμHLO of ~0.3%.  With such a large rate it will be necessary to develop a new triggerless front-end electronics, capable of reading out at 40 MHz.

This PhD project will focus on performing quality tracking in the real time trigger reconstruction. Algorithms of track finding techniques based on machine learning will be developed and applied to the reconstruction process at the online and offline stages.

You will be provided with comprehensive training in data science through LIV.INNO’s structured training program, as well as courses on theoretical and experimental particle physics. A 6-months industry placements will complement your training.