First Principle Simulator of a Stochastically Varying Image Plane for Photon-Counting High Contrast Applications

TL;DR

This paper introduces MEDIS, a comprehensive simulator for MKID-based high-contrast imaging, demonstrating how detector properties influence exoplanet imaging performance and identifying key factors for improvement.

Contribution

The paper presents MEDIS, a novel end-to-end simulator for MKID-based high-contrast imaging, incorporating degradation effects and analyzing performance impacts of detector properties.

Findings

Increasing pixel yield from 80% to 100% improves contrast by up to 8 times.

Enhancing maximum count rate and pixel sampling improves close-separation contrast.

Photon arrival time statistics can further improve low-flux imaging performance.

Abstract

Optical and near-infrared Microwave Kinetic Inductance Detectors, or MKIDs, are low-temperature detectors with inherent spectral resolution that are able to instantly register individual photons with potentially no false counts or readout noise. These properties make MKIDs transformative for exoplanet direct imaging by enabling photon-statistics-based planet-discrimination techniques as well as performing conventional noise-subtraction techniques on shorter timescales. These detectors are in the process of rapid development, and as such, the full extent of their performance enhancing potential has not yet be quantified. MKID Exoplanet Direct Imaging Simulator, or MEDIS, is a general-purpose end-to-end numerical simulator for high-contrast observations with MKIDs. The simulator exploits current optical propagation libraries and augments them with a new MKIDs simulation module to provide a pragmatic model of many of the degradation effects present during the detection process. We use MEDIS to demonstrate how changes in various MKID properties affect the contrast-separation performance when conventional differential imaging techniques are applied to low-flux, short duration observations. We show that to improve performance at close separations will require increasing the maximum count rate or pixel sampling when there is high residual flux after the coronagraph. We predict that taking pixel yield from the value achieved by current instruments of 80% and increasing it to 100% would result in an improvement in contrast of a factor of $\sim$ 4 at 3$\lambda/D$ and $\sim$ 8 at 6$\lambda/D$. Achieving better contrast performance in this low flux regime would then require exploiting the information encoded in the photon arrival time statistics.