Reconstruction of Beam Parameters and Betatron Radiation Spectra Measured with a Compton spectrometer

This project investigates the reconstruction of beam parameters from betatron radiation spectra in plasma wakefield acceleration (PWFA) experiments using a combination of simulated data and experimental modeling. Central to the approach is a custom Python-based tracking code that models radiation emission in the PWFA blowout regime by calculating spectra from Liénard-Wiechert fields. The analysis includes both one-dimensional and double-differential representations of the radiation. To infer beam properties such as spot size, emittance, and energy, the project implements a simulation-driven Maximum Likelihood Estimation (MLE) framework alongside machine learning (ML) models based on densely connected neural networks. These include spectral reconstruction using Expectation-Maximization (EM) algorithms and ML techniques applied to both 1D and 2D radiation spectra, including image-based models that use angular distributions as input. A key innovation in this work is the application of spectral normalization techniques, particularly for improving prediction accuracy at small spot sizes. The modeling and diagnostic tools are validated against Geant4 simulations and designed to align with realistic FACET-II constraints, enhancing their relevance for current and future experiments.

Machine learning-based analysis of experimental electron beams and gamma energy distributions(https://arxiv.org/abs/2209.12119)