🔭 astrophysics

Mathematical derivation and verification of the amplitude of LISA's interferometric signals on an ultra-stable interferometer testbed

This paper presents and validates an analytical framework that quantifies how beam tilts and wavefront curvature mismatches degrade interferometric signal amplitude and amplify phase noise in LISA-like testbeds, providing essential models for optimizing future gravitational wave detection systems.

Original authors: Alvise Pizzella, Lennart Wissel, Miguel Dovale-Alvarez, Pablo Martinez Cano, Rodrigo Garcia Alvarez, Christoph Bode, Juan Jose Esteban Delgado, Gerhard Heinzel

Published 2026-07-03

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Alvise Pizzella, Lennart Wissel, Miguel Dovale-Alvarez, Pablo Martinez Cano, Rodrigo Garcia Alvarez, Christoph Bode, Juan Jose Esteban Delgado, Gerhard Heinzel

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: Listening to the Universe with Lasers

Imagine the LISA mission as a giant, floating ruler in space, designed to measure the distance between three spacecraft with incredible precision—down to the width of a single atom (picometers). To do this, they shoot laser beams between the spacecraft.

However, space isn't perfectly still. The spacecraft wobble and tilt slightly. When a laser beam hits a detector (a special camera called a Quadrant Photodiode or QPD) at a slight angle, the measurement gets "fuzzy." This fuzziness is called noise. The paper investigates exactly how much this fuzziness increases when the laser beam tilts, and how to predict it mathematically.

The Core Problem: The "Flashlight" Analogy

Think of the laser beam like a flashlight shining on a wall.

Perfect Alignment: If you shine the flashlight straight on the wall, the light is bright and centered. The detector sees a strong, clear signal.
Tilted Beam: If you tilt the flashlight, the light spreads out, hits the wall at an angle, and might not cover the center of the detector as well. The signal gets dimmer.
The "Wavefront" Mismatch: Now, imagine the flashlight beam isn't just a straight cylinder; it's slightly curved (like a bowl) or shaped differently than the detector expects. If you tilt a curved beam, the way the light hits the detector becomes uneven. One side of the detector might get a lot of light, while the other gets very little.

The paper's main discovery is that this unevenness (caused by the mismatch in shape) makes the noise worse than we previously thought.

Key Findings Explained

1. The "Split Screen" Effect (The QPD)

The detectors LISA uses aren't just one big sensor; they are like a four-pane window (a Quadrant Photodiode). The computer looks at the light hitting the top two panes versus the bottom two panes to figure out if the beam is tilting.

The Old Assumption: Scientists used to think that if the beam tilted, the signal would just get a little weaker, but the top and bottom panes would behave the same way.
The New Discovery: The paper shows that if the laser beams have a slight "shape mismatch" (curvature mismatch), tilting the beam makes the top panes and bottom panes behave very differently. One side might stay bright while the other gets dim.
The Result: Because the two sides are now doing different things, the computer gets confused, and the "noise" (static) in the measurement goes up. The paper provides a new mathematical formula to predict exactly how much the noise increases based on how "mismatched" the beams are.

2. The "Infinite vs. Real" Detector

The authors first did the math for a detector that is infinitely large (like a wall that goes on forever). This was easy to calculate. Then, they did the math for a real, finite-sized detector (the actual 1mm size used in the lab).

The Finding: They found that for small detectors, the "shape mismatch" effect is actually less severe than for huge detectors. It's like trying to fit a large, wavy blanket onto a small table; the table just cuts off the messy edges, so the problem is smaller.

3. The Lab Test (The "Testbed")

To prove their math wasn't just theory, the authors built a miniature version of the LISA system in a lab in Hannover, Germany.

They used a super-stable table (the "testbed") to simulate the spacecraft.
They tilted the laser beams and measured the signal.
The Verdict: The real-world measurements matched their new mathematical formulas perfectly. This confirms that their new model is accurate.

4. Should We Tilt the Mirror on Purpose?

In the LISA mission, sometimes the spacecraft are built with tiny imperfections, causing the laser to hit the detector at a slight angle naturally.

The Dilemma: One idea was to intentionally tilt the test mass (a mirror inside the spacecraft) to "fix" the angle and make the laser hit straight on again. This would make the signal stronger (less noise from the signal itself).
The Trade-off: However, tilting the mirror introduces a different kind of noise (called "Tilt-to-Length" coupling), where the mirror's wobble gets mixed into the distance measurement.
The Conclusion: The authors calculated that the tiny gain in signal quality from fixing the angle is not worth the extra noise introduced by tilting the mirror. It is better to leave the mirror straight and accept the tiny signal loss.

Summary of the "New Feature"

The paper highlights a specific, previously unnoticed feature: When laser beams have a shape mismatch and are tilted, the noise doesn't just go up uniformly; it becomes "lopsided."

Imagine a seesaw. If you push down on one side (tilt), and the seesaw is perfectly balanced, it goes down evenly. But if the seesaw is slightly warped (shape mismatch), pushing down on one side makes the other side jump up wildly. This "lopsided" behavior creates extra static in the measurement. The paper provides the tools to calculate exactly how much static this creates.

Why This Matters

This work gives the LISA mission a "rulebook" for understanding how much noise to expect when the spacecraft wobbles. It tells engineers that they don't need to worry about tilting the mirrors to fix small alignment errors, because the math shows it won't help enough to be worth the risk. It also ensures that the mission's data analysis software accounts for this new "lopsided" noise effect to keep the gravitational wave measurements as clean as possible.

Technical Summary: Mathematical Derivation and Verification of the Amplitude of LISA's Interferometric Signals

Problem Statement
The Laser Interferometer Space Antenna (LISA) mission aims to detect gravitational waves by measuring the separation between free-falling test masses (TMs) with picometer-level sensitivity ( $\text{pm}/\sqrt{\text{Hz}}$ ). A critical challenge in achieving this sensitivity is the presence of beam tilts at the photodetectors, caused by spacecraft and test mass angular jitter. These tilts degrade the heterodyne efficiency (HE)—the overlap between the interfering beams—thereby reducing the signal amplitude and the signal-to-noise ratio (SNR). A reduced SNR increases the coupling of additive readout noise (such as shot noise and electronic noise) into the phase measurement. While previous work established the relationship between beam tilt and the Differential Wavefront Sensing (DWS) signal, a comprehensive analytical model describing the tilt-dependent amplitude response (HE) for finite detectors, particularly Quadrant Photodiodes (QPDs) with wavefront curvature mismatches, was lacking. Furthermore, the impact of these amplitude variations on the total phase noise budget, specifically distinguishing between correlated and uncorrelated noise sources, required rigorous quantification.

Methodology
The authors developed an analytical framework based on minimal, physically justified approximations to relate beam tilt to the heterodyne signal amplitude. The study considers two primary interferometric topologies relevant to LISA:

Test Mass Interferometer (TMI): Interference between two fundamental-mode Gaussian beams (GB-GB).
Science Interferometer (SCI): Interference between a Gaussian beam and a "flat-top" (top-hat) beam (GB-TH), resulting from the propagation of a Gaussian beam over interplanetary distances and subsequent clipping.

The analysis covers two detector geometries:

Single-Element Photodiodes (SEPDs): Used for theoretical completeness and comparison.
Quadrant Photodiodes (QPDs): The actual detectors used in LISA, modeled as slit-less segments for analytical tractability, with extensions to finite sizes.

The derivation proceeds by:

Defining the electric fields of the reference and measurement beams in a laboratory reference frame, accounting for beam rotation around the detector center.
Calculating the overlap integral of the beam intensities over the detector surface to determine the heterodyne power term.
Deriving exact solutions for infinite detectors and closed-form Maclaurin series approximations (up to seventh order) for finite detectors.
Validating these analytical expressions against numerical simulations (using a custom integrator) and experimental measurements.
Utilizing the derived HE models to quantify the tilt-dependent coupling of phase noise, distinguishing between uncorrelated noises (shot noise, electronic noise) and correlated noises (Relative Intensity Noise - RIN).

The experimental validation was performed on the Tilt-to-Length (TTL) coupling and DWS Optical Bench Simulator (TDOBS), an ultra-stable testbed representative of the LISA optical bench. The setup utilized a 1064 nm Nd:YAG laser, acousto-optic modulators for frequency shifting, and commercial QPDs with associated trans-impedance amplifiers.

Key Contributions

Analytical HE Models: The paper provides exact analytical expressions for the heterodyne efficiency of infinite SEPDs and QPDs, and Maclaurin series approximations for finite detectors. These models explicitly incorporate the effective spot radius, wavefront curvature mismatch ( $\rho$ ), and tilt angle ( $\theta$ ).
Discovery of QPD Asymmetry: A novel finding is that in the presence of wavefront curvature mismatch ( $\rho \neq 0$ ), the heterodyne efficiency of individual QPD segments becomes asymmetric with respect to the tilt angle. Specifically, the peak HE for the top and bottom segments shifts to non-zero, opposite tilt angles. This effect was previously unmodeled in the context of LISA's readout noise.
Noise Coupling Factors: The authors derive new coupling factors ( $k_{unc}$ , $k_{cor,\Sigma}$ , $k_{cor,\Delta}$ ) that capture the angular dependence of uncorrelated and correlated phase noises in QPD readouts. These factors demonstrate that treating the QPD as a single entity (or assuming symmetric segment response) is insufficient when wavefront curvature mismatches are present, as the noise contribution from segments with lower HE dominates.
Experimental Verification: The analytical models were validated against high-precision measurements on the TDOBS testbed. The results confirmed the predicted asymmetry in HE peaks for QPD halves and the general dependence of HE on tilt and wavefront mismatch.
Impact on LISA Noise Budget: The study quantifies the noise penalty associated with operating the test masses at small static tilts (to correct for optical bench misalignments, a strategy known as "DWS-zero"). The analysis suggests that while tilting improves HE, the resulting noise reduction is marginal (order of $10^{-3}$ ) compared to the potential increase in TTL coupling noise, suggesting that operating with a DWS offset may be preferable.

Results

HE Degradation: Beam tilts significantly reduce HE. For a typical LISA configuration, a tilt of 1 mrad can reduce HE by a factor of up to five.
Wavefront Curvature Mismatch: The parameter $\rho$ (normalized wavefront curvature mismatch) is a critical factor. Even at zero tilt, a non-zero $\rho$ reduces the maximum achievable HE. More importantly, it causes the HE peak to shift away from zero tilt for individual QPD segments, creating an asymmetry that increases the overall phase noise coupling.
Noise Increase: In the worst-case scenario (maximum expected $\rho$ in LISA), the presence of wavefront curvature mismatch increases the uncorrelated readout noise by approximately 10-16% even without tilts. With tilts, the noise coupling increases further.
Experimental Agreement: The experimental measurements of normalized HE versus tilt angle on the TDOBS testbed showed excellent agreement with the numerical models and the analytical approximations, particularly in capturing the asymmetry between QPD halves.
Static Tilt Trade-off: Calculations for the TMI indicate that intentionally tilting the test mass to compensate for LOB misalignments (to achieve DWS-zero) yields a very modest reduction in interferometric noise (relative increase of $\sim 10^{-3}$ to $10^{-2}$ ). This small gain is likely outweighed by the increased coupling of the test mass's lateral jitter to the path length signal (TTL noise).

Significance and Claims
The paper claims to provide the first analytical description of the heterodyne efficiency for QPDs that accounts for the asymmetry induced by wavefront curvature mismatches. This is significant because it corrects a potential oversight in noise budgeting where QPD segments were assumed to have identical responses. The authors emphasize that for accurate phase noise modeling in LISA, the QPD must be treated as a whole, and individual segment behaviors must be considered when wavefront mismatches are present.

The derived models are general and applicable to any interferometric topology subject to tilts around the detector plane, though they are specifically tailored to the constraints of space-based interferometers like LISA, Taiji, and Tianqin. The work provides a compact function to capture the angular dependence of correlated and uncorrelated phase noises, offering a tool for optimizing the design and operation of future gravitational wave missions. The authors modestly conclude that while their models predict a noise benefit from correcting misalignments via static tilt, the net benefit is likely negligible compared to the TTL noise penalty, a conclusion supported by the experimental data and pending further publication on DWS performance.