Upgrade of the MITHiC testbed for phase and amplitude control in the focal plane
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Astrophysics. The day in astrophysics: observations, models and what the data forces.
Vera: I'm Vera.
Jocelyn: And I'm Jocelyn, here with Subrahmanyan.
Vera: We're looking at the paper: "Upgrade of the MITHiC testbed for phase and amplitude control in the focal plane".
Jocelyn: Here's the gist of the paper. High-contrast imaging for the direct detection of exoplanets is limited by stellar brightness and by wavefront errors.
Vera: Within this framework, the MITHiC testbed at the Laboratoire d’Astrophysique de Marseille is being upgraded with a new deformable mirror used to control the wavefront.
Jocelyn: This contribution presents recent progress on the MITHiC upgrade and initial results, highlighting its role as a versatile platform for developing advanced high-contrast imaging techniques.
Vera: Testbed Overview MITHiC is a visible-light optical bench at LAM designed to test NCPA correction methods and high-contrast coronagraphic imaging algorithms, including dark-hole techniques such as COFFEE8 and nonlinear dark-hole algorithms.
Title and authors: Vera: The paper is titled "Upgrade of the MITHiC testbed for phase and amplitude control in the focal plane". It's a straightforward description of what they did on this optical bench.
Jocelyn: And it’s important because it shows how this specific hardware, the MITHiC testbed, can be made more versatile for developing advanced high-contrast imaging techniques.
Subrahmanyan: This upgrade is significant because it moves beyond just testing a single algorithm and establishes a platform that can handle various correction methods.
Vera: It’s about showing that this bench isn't just one thing; it can be adapted to test different kinds of coronagraphic imaging algorithms.
Jocelyn: They are highlighting its role as a versatile platform, which is important because it means we can develop new techniques there without having to build entirely new hardware for every test.
Subrahmanyan: So, the implication is that this upgrade accelerates the development cycle for high-contrast imaging methods by providing a flexible and capable testing ground.
The paper's summary: Vera: The paper summarizes how they implemented this upgrade, which involves adding a new deformable mirror to control the wavefront. They also bring in a ZELDA sensor for wavefront sensing.
Jocelyn: The ZELDA sensor is crucial because it provides a direct pixel-by-pixel estimate of the optical path difference, or OPD, in nanometers.
Subrahmanyan: This direct measurement capability is what allows them to implement an NCPA correction loop without needing an interaction matrix for the correction pipeline.
Vera: That’s a big point because it simplifies the correction process; they reconstruct the OPD using a specific equation and then correct for geometric distortion between planes using an affine transformation.
Jocelyn: They also filtered that correction in Fourier space using a top-hat filter, which restricts the corrections to a useful band where the residual root mean square error decreases until about seventy-five times per pixel.
Subrahmanyan: The summary emphasizes that they characterize the nominal bench performance and then move on to characterizing the radial contrast profile and finally looking at the photon-noise limit.
The paper's improvements: Vera: One key improvement is the NCPA correction loop itself, which they describe as operating without an interaction matrix because of that direct pixel-by-pixel OPD estimate from ZELDA.
Jocelyn: That loop involves acquiring background, reference, and ZELDA frames to reconstruct the OPD using pyZELDA, and then correcting for geometric distortion between the spatial light modulator and the ZELDA pupil planes.
Subrahmanyan: They also introduce a Gaussian deformable-mirror simulation framework to emulate a real deformable mirror with twenty-seven four by two seven four actuators directly on the bench without needing any hardware modifications.
Vera: That simulation helps them define what we call the correction floor, or sigma fit, which they find depends only on the bench OPD and the chosen actuator basis, not on how many iterations they run or how much loop gain they use.
Jocelyn: The simulation suggests that for a specific configuration of a twenty-seven four by two seven four Gaussian deformable mirror spanning the full pupil with a coupling of zero point two zero, the estimated floor is about six point one nanometers.
Conclusion: Vera: So, to wrap up this paper on the "Upgrade of the MITHiC testbed for phase and amplitude control in the focal plane", they show that they've made substantial progress in controlling wavefront errors.
Jocelyn: The main implication is that this upgraded platform is now a more versatile tool for testing various high-contrast imaging algorithms, thanks to features like the ZELDA sensor and the new deformable mirror.
Subrahmanyan: From a cosmology perspective, it means we have a more robust laboratory to test the methods that will eventually be applied when we look at faint stars across the universe.
Vera: The photon-noise limit analysis they do also helps us understand where we are limited by noise versus where we are limited by speckles, showing a crossover radius of about eleven lambda over diameter after the correction is applied.
Jocelyn: It’s exciting because it shows how careful control over these aberrations can push the limits of what we can actually measure in direct exoplanet detection.
Subrahmanyan: Overall, this work solidifies MITHiC as a strong platform for testing algorithms that aim to reach the contrast levels needed for finding faint planets around stars.
Anton Neisiusa, Élodie Choqueta, Vincent Chambouleyrona, Arthur Vigana
Aix Marseille University · CNRS · LAM
astro-ph.IM
Submitted: 2026-11-09
Updated: 2026-11-09
Comments: 12 pages, 10 figures, SPIE Astronomical Telescopes + Instrumentation 2026, Copenhagen
Journal ref: Proc. SPIE 14145, 1414565 (2026)
DOI: 10.1117/12.3103968
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Importance score: 87/100
The gist: High-contrast imaging for the direct detection of exoplanets is limited by stellar brightness and by wavefront errors.
Key concepts
- MITHiC Testbed
- MITHiC is a visible-light optical bench at LAM used to test methods for correcting wavefront errors and coronagraphic imaging algorithms. It simulates observing a point star using a single laser diode at 670.7 nm, allowing researchers to develop tools for detecting faint exoplanets.
- NCPA Correction Loop
- This is a pipeline that estimates and corrects optical path difference (OPD) errors pixel by pixel using data from the ZELDA sensor. It uses a specific mathematical reconstruction method to correct geometric distortions between the mirror and sensor planes, filtering corrections in Fourier space to stabilize performance.
- Dark-hole Techniques
- These are specific coronagraphic algorithms, like COFFEE8 and nonlinear dark-hole methods, used in high-contrast imaging. They aim to suppress the light from a bright central star by creating a 'dark hole' or region of suppressed contrast where planets can be seen.
- Gaussian Deformable Mirror Simulation
- A simulation framework that mimics the behavior of a real deformable mirror with N2 actuators without needing physical hardware changes. It allows researchers to predict performance metrics, such as the correction floor, based on the simulated mirror's properties.
Summary
High-contrast imaging for the direct detection of exoplanets is limited by stellar brightness and by wavefront errors. p. 0 Within this framework, the MITHiC testbed at the Laboratoire d’Astrophysique de Marseille is being upgraded with a new deformable mirror used to control the wavefront. p. 0 This contribution presents recent progress on the MITHiC upgrade and initial results, highlighting its role as a versatile platform for developing advanced high-contrast imaging techniques. p. 0
Testbed Overview
MITHiC is a visible-light optical bench at LAM designed to test NCPA correction methods and high-contrast coronagraphic imaging algorithms, including dark-hole techniques such as COFFEE8 and nonlinear dark-hole algorithms. The setup uses a single-mode laser diode at λ = 670. p. 17 nm simulating a point star. p. 1 Wavefront sensing is performed by a ZELDA sensor, which consists of a focal-plane phase mask combined with a pupil-plane camera to reconstruct the optical path difference (OPD). p. 1 The science camera records the PSF and coronagraphic images at a gain of gcam = 2. p. 101 e−/ADU. p. 1
NCPA Correction Loop
The NCPA correction pipeline operates without an interaction matrix because the ZELDA sensor provides a direct pixel-by-pixel estimate of the OPD in nanometers. The loop involves acquiring background, reference, and ZELDA frames to reconstruct the OPD via Eq. p. 3 (1) using pyZELDA. p. 3 The measured OPD is then corrected for geometric distortion between the SLM and ZELDA pupil planes using an affine transformation calibrated at the beginning of the session. p. 3 The correction is filtered in Fourier space using a top-hat filter, restricting corrections to a useful band where residual rms decreases monotonically until νc = 75 c/p. p. 3
Performance Characterization
The nominal performance showed the initial OPD was measured at 34.8 nm rms, and after NCPA correction, it dropped to 2. p. 24 nm rms in 15 iterations. p. 3 The photon-noise limit is derived from first principles as Equation (8), which shows the contrast is limited by both the dilution of noise over a growing number of pixels and the radial decay of the PSF itself. p. 5 The crossover radius rcross separating speckle-limited and photon-noise-limited regimes is approximately 11 λ/D after NCPA correction. p. 5
Simulation Framework
A Gaussian deformable-mirror simulation framework is introduced to emulate the behavior of a real DM with N2 actuators directly on the bench without hardware modification. The correction floor, σfit, is defined by the Pythagorean decomposition of orthogonal projection as Equation (11). p. 6 This floor depends only on the bench OPD and the chosen actuator basis, not on loop gain or iteration count. p. 6 The simulation suggests that for a 5 × 5 Gaussian DM with coupling c = 0. p. 620 spanning the full pupil, the estimated floor is approximately 6. p. 71 nm.
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Improvement of NCPA correction loop: The system can implement a ZELDA-based control loop that applies an arbitrary phase map to a spatial light modulator to correct low-order aberrations without needing an interaction matrix, as described in Section 3.1.
-
Improvement of DM emulation for dark hole characterization: The AI system can utilize the Gaussian influence function model to map the function Cdarkhole(N, c), allowing it to predict the optimal actuator count and coupling coefficient needed to minimize dark-hole contrast based on a single OPD measurement.
-
Improvement of open-loop predictive correction: By integrating thermal monitoring data with Zernike coefficient sensitivities (e.g., ∂aZ2/∂T = +20.9 nm/◦C), the AI system can apply open-loop predictive corrections from temperature readings alone to maintain dark-hole stability over long exposures without continuous ZELDA measurements.
-
Improvement of noise limit analysis: The system can calculate the photonnoise normalized intensity limit Cphoton(r) using a first principles derivation, allowing it to identify the crossover radius rcross separating the speckle-limited regime from the photon-noise limited regime.
Related papers
- On-sky demonstration of a vector Zernike wavefront sensor in a cascaded adaptive optics system
- Euclid Data Release 1 (DR1): Calibration strategy and observations
- The General Antiparticle Spectrometer (GAPS) Antarctic Balloon Payload
- Detection and excision of radio-frequency interference in radio arrays
- Unified matched-filter and excess-power pipeline for gravitational-wave detector characterization
- Ushering in sub-arcsecond angular resolution observatories in the far-infrared: The case for a double-Fourier interferometer