Microcalorimeter pulse analysis by means of principle component decomposition

TL;DR

This paper explores using principal component analysis for microcalorimeter pulse analysis, comparing it to traditional methods and assessing effects of instrumental factors through simulations and real data application.

Contribution

It introduces PCA as an alternative to optimal filtering for pulse analysis in microcalorimeters, including simulation studies and real data application.

Findings

PCA effectively describes pulse data with main components.

Instrumental effects influence PCA analysis results.

Comparison shows PCA's potential advantages over traditional methods.

Abstract

The X-ray integral field unit for the Athena mission consists of a microcalorimeter transition edge sensor pixel array. Incoming photons generate pulses which are analyzed in terms of energy, in order to assemble the X-ray spectrum. Usually this is done by means of optimal filtering in either time or frequency domain. In this paper we investigate an alternative method by means of principle component analysis. This method attempts to find the main components of an orthogonal set of functions to describe the data. We show, based on simulations, what the influence of various instrumental effects is on this type of analysis. We compare analyses both in time and frequency domain. Finally we apply these analyses on real data, obtained via frequency domain multiplexing readout.