PCA-based Data Reduction and Signal Separation Techniques for James-Webb Space Telescope Data Processing

TL;DR

This paper reviews PCA-based and related techniques like ICA, NMF, and SSA for processing James Webb Space Telescope data, aiming to improve data reduction and signal separation in astronomical imaging.

Contribution

It provides a comprehensive overview of PCA and similar methods applicable to JWST data processing, highlighting potential pipeline interventions.

Findings

PCA techniques effectively separate uncorrelated data components.

ICA can identify independent signal sources in JWST data.

NMF and SSA offer interpretable and time-series analysis methods.

Abstract

Principal Component Analysis (PCA)-based techniques can separate data into different uncorrelated components and facilitate the statistical analysis as a pre-processing step. Independent Component Analysis (ICA) can separate statistically independent signal sources through a non-parametric and iterative algorithm. Non-negative matrix factorization is another PCA-similar approach to categorizing dimensions in physically-interpretable groups. Singular spectrum analysis (SSA) is a time-series-related PCA-like algorithm. After an introduction and a literature review on processing JWST data from the Near-Infrared Camera (NIRCam) and Mid-Infrared Instrument (MIRI), potential parts to intervene in the James Webb Space Telescope imaging data reduction pipeline will be discussed.