CMA-unfold: A covariance matrix adaptation unfolding algorithm for stacked calorimeter detectors

Area of Science:

• Nuclear Physics
• Plasma Physics
• Experimental Diagnostics

Background:

• Stacking calorimeters are crucial for measuring high-energy particles in fusion and laser experiments.
• Analyzing this data involves solving complex inverse problems with noise and detector uncertainties.

Purpose of the Study:

• To develop an open-source unfolding framework for accurate photon spectrum reconstruction.
• To provide a robust and flexible tool for stacking calorimeter data analysis.

Main Methods:

• Introduced an open-source framework (ggfauvel/CMA-unfold) based on covariance matrix adaptation evolution strategy.
• Reconstructed photon spectra directly from depth-dose profiles without parametric assumptions.

Main Results:

• The algorithm demonstrated high robustness in recovering complex spectral shapes.
• It accurately tolerated percent-level deviations in detector layers, showing noise resilience.

Conclusions:

• The framework offers a flexible and noise-resilient solution for stacking calorimeter data.
• It is particularly relevant for bremsstrahlung diagnostics in high-intensity laser and inertial confinement fusion.