University of Leicester
Browse

Can machine learning reduce the number of anode readouts for reconstruction of coincident single photons in CDIR resistive sea photon detectors?

journal contribution
posted on 2025-01-23, 12:01 authored by A Markfort, Alexander BaranovAlexander Baranov, TM Conneely, A Duran, Jonathan LapingtonJonathan Lapington, J Milnes, W Oughton, I Tyukin
This study focuses on exploring the potential of Charge Division Imaging Readout (CDIR) for micro-Channel plate (MCP) based resistive sea photon detectors. The CDIR technique spreads the MCP charge footprint capacitively between readout nodes forming anode segments. Charge measurements at each node are then used to reconstruct incident photon's position and time. A primary objective is to investigate the minimum number of anode segmentation's necessary, to allow successful reconstruction of multiple photons within a given time interval where pile up would be an issue for traditional approaches. Allowing for optimisation of the anode structure, investigating for a readout schematic to improve timing, rate capability, and reduce distortion effects. Algorithmic and machine learning (ML) techniques will be compared and utilised to reconstruct spatial positions of multiple photons. The comparison will aim to determine if machine learning techniques can be utilised to correct for algorithmic systematic errors to provide a more robust system, whilst removing the need for complex calibrations and allowing for efficient implementation on FPGA in future work.

Funding

1851 Royal Commission for funding this PhD and the UKRI Turing AI Fellowship EP/V025295/1

History

Author affiliation

College of Science & Engineering Comp' & Math' Sciences Physics & Astronomy

Version

  • AM (Accepted Manuscript)

Published in

Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment

Volume

1069

Pagination

169863

Publisher

Elsevier BV

issn

0168-9002

eissn

1872-9576

Copyright date

2024

Available date

2025-01-23

Language

en

Deposited by

Professor Jon Lapington

Deposit date

2024-12-02