← Back to Main Page

Making a Small Lab Act Infinite: Acoustic Cloaking, Holography, and Cloning

If you have ever clapped in a small room or dropped a pebble in a bucket, you know the problem: waves bounce off boundaries. In an open ocean, a quiet field, or outer space, waves can radiate away as if the world extends forever. In a laboratory, they hit walls and come right back. Those echoes contaminate measurements, shrink the region where results can be trusted, and make it harder to study how waves behave in the wild.

We decided not to fight that limitation with bigger rooms or thicker foam. Instead, we taught the walls to help. The approach blends a physical experiment with a virtual world in real time through Immersive Boundary Conditions (IBCs). The lab stays small, but the waves inside it behave as though the outside continues beyond the walls. With IBCs we can suppress wall reflections, make objects acoustically invisible, project virtual scatterers into a real experiment, and even clone the scattering fingerprint of an object on demand across a broad range of frequencies. For more details, see Centre for Immersive Wave Experimentation

Related: Landrø & Favretto-Cristini, Acoustic Physics (2018).

1) The challenge: small labs, bouncing waves

In the wild, waves in open water, air, or the ground can radiate away and leave the scene. In tanks, rooms, and waveguides, they ricochet off walls, polluting measurements and burying subtle effects such as weak scattering and fine phase shifts. Anechoic chambers help, but their absorbers must scale with wavelength, so low-frequency experiments quickly become impractically large. In solid 3D volumes for elastic P- and S-wave studies, wavelengths around 1-20 kHz can approach the sample size, allowing boundary energy to dominate the interior wavefield. That scale gap makes it difficult to transfer laboratory insight to field problems, especially for frequency-dependent phenomena such as attenuation, dispersion, anisotropy, and nonlinearity.

Related: Xun Li, Elastic immersive wave experimentation – Doctoral Thesis, ETH Zurich, 2022.

2) Immersive Boundary Conditions: Active wave energy cancellation

In size-limited laboratories, reflections from rigid boundaries often obscure the waves scattered by interior objects. One elegant solution is the deployment of active sources around the physical experimental domain. In an ideal implementation, the active boundary sources generate counter-waves that completely cancel the outgoing wavefield. As a result, the only waves visible in the experiment are those related to the interior scatterers, free from unwanted boundary effects. This clear separation is crucial for accurate measurement and analysis. Also, in this way, the experiment is virtually "immersed" into a much larger space where the propagation characteristics are as if no rigid boundaries were present. In effect, the physical wave propagation mirrors the behavior of waves in an idealized, boundary-free environment.

The active sources can create a virtual environment by emitting waves designed not only to cancel naturally outgoing waves through carefully controlled emissions, but also to actively represent interactions between a physical experiment and an artificial virtual domain. In this configuration, the physical wave propagation experiment becomes seamlessly connected to an extended or virtual environment—the emitted waves embodying the desired properties of an unbounded, immersive domain while the propagating waves remain explicit in the physical realm.

Simulation movies of this approach strikingly illustrate that, with active cancellation, the laboratory’s confined physical space transforms into an immersive experimental domain. This environment faithfully replicates real-world scattering phenomena and supports broadband measurements without the distortions typically caused by reflections.

3) Immersive Boundary Conditions (IBCs): Real-time, broadband acoustic implementation

IBCs let us virtually replace what lies beyond a control surface by extrapolating measurements from an outer recording ring to an inner emitting ring in real time. In our implementation, pressure and particle velocity measured on a closed outer surface are used to predict the incident wave field on the inner boundary using representation theorems together with precomputed Green’s functions as “wave kernels” to extrapolate the field. This approach provides deterministic, broadband, and global control at low latency, independent of frequency or incident angle. With low-latency FPGAs (no more than 200 μs compute time), the extrapolated wave field is immediately re-injected into the lab so that waves cross seamlessly between the physical and virtual domains. This real-time prediction method cancels unwanted boundary reflections, enabling complex effects such as cloaking and holography without any prior knowledge of the incident wave.

Related: Becker et al., PRX, 2018.

4) Cloaking: Perfect invisibility at all angles without prior source knowledge

Think Harry Potter’s invisibility cloak: light bends around him and no one sees a thing. Doing that for light in open space is famously hard—an optical cloak would need to steer every color and every angle around an object and still deliver the wavefront to the observer “on time.” Passive materials run into causality and dispersion limits, so practical optical cloaks tend to be narrowband, angle/polarization‑limited, or confined to special geometries.

For sound, we take a different route: instead of hiding the object, we hide its acoustic signature. Using interior boundary control (IBC), an outer ring of microphones listens and a real‑time extrapolation predicts the wavefield at an inner ring of loudspeakers. Those speakers emit a secondary field that cancels the object’s scattering—including multipath from the lab boundaries—without any prior model of the incident sound. Even moving or unknown sources can be cloaked.

What does “invisible” look like in practice? We cloaked a quasi‑rigid circular scatterer (diameter 12.6 cm) inside a 2D waveguide by surrounding it with 20 control loudspeakers. A moving broadband primary field was synthesized by eight sources along a 96° arc. With the cloak off, both back‑ and forward‑scattering are visible; with the cloak on, the measured total field matches the “no‑object” reference—even at later times when wall echoes dominate. Finite‑element simulations closely reproduce the measurements. Across ~3.5 octaves (up to ~8.7 kHz), the mean scattered intensity at the outer array is reduced by about −8.4 dB and the angular scattering pattern is uniformly suppressed, demonstrating all‑angle, source‑agnostic cloaking.

Examples adapted from Becker et al., Science Advances, 2021.

5) Holography: Cheat an observer with virtual imprints

Picture the classic sci‑fi hologram: a lifelike person appears in front of you to talk, even though they’re far away. The trick is that your eyes receive the exact wavefronts they would have seen if that person were really there. Acoustic holography does the same for sound. We make microphones and loudspeakers conspire so an observer hears precisely the scattered field a chosen object would produce—even when that object isn’t present. By predicting, in real time, the wavefield on a sound‑transparent inner surface and driving collocated monopole+dipole emission (two close loudspeaker rings), we “imprint” a virtual object into the lab. To microphones (and your ears), the illusion is indistinguishable from the real scatterer. Note that we use the same real-time extrapolation as for cloaking, but now to recreate the object’s scattered field rather than cancel it.

Examples adapted from Becker et al., Science Advances, 2021.

See also: van Manen et al., JASA, 2015 (original theory); Börsing et al., Phys. Rev. Applied, 2019 (1D demonstration).

6) Acoustic Cloning: A Digital Twin That Scatters Like the Real Object

Acoustic cloning creates a digital twin of a real scatterer: not a copy of its material, but a replayable copy of how it scatters sound. The idea is to record the object's true scattering response in a controlled experiment, retrieve a clean version of that response, and then use it to "play back" the object's influence on any incident wavefield.

• Retrieval of the real scatterer’s properties through Multidimensional Deconvolution (MDD).
• Holographic reconstruction of the scatterer’s imprint via immersive holographic techniques.

Step 1: Acquiring the Scatterer’s Green’s Functions Using MDD

The first stage illuminates the real scatterer with controlled broadband signals. During this illumination, wavefields are recorded on a closed, sound-transparent receiver aperture surrounding the scatterer. Multidimensional deconvolution (MDD) is then used to extract the scatterer's Green's functions: the transfer responses that describe how sound propagates from the sources, interacts with the object, and scatters back through the medium. Crucially, MDD removes unwanted contributions such as multiple reflections and boundary effects that would otherwise pollute the object's clean scattering signature.

By eliminating the boundary‑induced reverberations, the resulting Green’s functions represent the object’s response under ideal radiation conditions. In other words, we retrieve the “pure” scattering characteristics as if the object were in an unbounded space.

Once the raw data have been decomposed into their incident and outgoing components (a key process in MDD), a set of coupled Fredholm integral equations is formulated. Solving these equations in a least‑squares sense enables us to reconstruct the Green’s functions that reveal the object’s scattering behavior.

What Does This Mean Practically?
No prior knowledge of the object’s material properties, geometry, or source signature is required. Every detail is captured from the measurement data, making the approach remarkably general and robust even in reverberant laboratory environments.

In summary, after illuminating the scatterer and applying MDD, we obtain a set of Green’s functions that isolate and capture the object’s scattering response. For example, by subtracting the homogeneous (direct wave) component, only the contributions from the scatterer remain.

Step 2: Reconstructing the Scatterer with Real‑Time Holography

With the scatterer's Green's functions in hand, the next step is to recreate the object's acoustic imprint in the lab, even after the physical object has been removed. This is achieved using a holographic reconstruction process:

1. Deactivate the Object: The physical scatterer is removed from the experimental domain.
2. Replay Green’s Functions: The previously retrieved Green’s functions are used as signal templates to drive an array of transducers on the inner ring (SI). These transducers emit time‑varying monopole and dipole signals derived from the Green’s functions.

The resulting wavefield, excited by these precisely calibrated signals, reproduces all scattering phenomena—including multiple interactions between the incident wave and the scatterer’s numerical imprint—exactly as if the original object were still in place.

A few notable features of the holographic clone:
Broadband robustness: the hologram responds correctly for incident broadband wavefields.
Versatility: whether the original scatterer is circular, square, or cross-shaped, the holographic reconstruction reproduces its scattering features.
Real-time reproduction: the reconstruction occurs with low latency, so echoes and interference patterns build up in real time.

Augmenting the Digital Twin: Beyond Cloning

An especially exciting aspect of this approach is that the twin is defined digitally. Once the scatterer’s Green’s functions have been retrieved, they can be manipulated arbitrarily. For instance, you can:

• Rotate or Translate the Clone: Configure different spatial arrangements without physically modifying the original object.
• Scale Amplitudes: Introduce directional gain to alter the amplitude of scattering in specific directions.
• Modify Impedance or Transparency: Create “acoustic cyborgs” that interact with real wavefields while exhibiting non‑physical (or enhanced) behaviors.

These digital modifications can be made prior to playback through the holographic emitter. This opens up applications in:

• Enhanced metamaterial designs, where unit cells can be cloned and arranged without the challenges of physical manufacturing.
• Virtual acoustic models, allowing simulation of an object’s response in various environments or configurations—thereby enabling rapid prototyping and testing.

Wandering Thoughts

Acoustic cloning is not a copy of atoms or materials. It is a copy of the way an object behaves when sound meets it: its scattering fingerprint. Once that fingerprint has been measured and reconstructed, the object can be removed while its acoustic response remains available for controlled experiments.

This matters because scattering is often the quantity we want to test. Acoustic cloning captures the interaction between a wavefield and an object across angles, frequencies, and multiple reflections. That makes it possible to prototype changes digitally after one carefully controlled measurement.

Possible applications include:

• Rapid design loops: Measure a unit cell once, clone it, modify it digitally, and test variants without repeated fabrication.
• Acoustic heritage and education: Preserve the “sound” of instruments, halls, or artifacts and let students interact with them anywhere.
• Personalized spaces: Car cabins, classrooms, and headsets that remix their acoustic behavior in real time for comfort, clarity, or immersion.
• Metamaterials, demystified: Clone a unit cell, tile it virtually, and explore nonphysical variants (gain, custom directivity) before ever building.

Of course, power invites responsibility. If we can project acoustic “truths” into a room, we should mark what’s real and what’s rendered. Standards for safe output levels, disclosure in public spaces, and reproducible reporting will help keep the tech trustworthy. And we should keep sight of the limits: a clone is only as good as the bandwidth and geometry it learned from.

What I find useful about this idea is that it stays experimental. The result is not only a simulation on a screen; it is a reconstructed wave interaction inside a physical lab. We can ask how the response changes if the object is rotated, softened, or made directional, and test those changes with real wavefields.

Examples adapted from Müller, J., Becker, T. S., Li, X., et al., Physical Review Applied (2023).

7) Applications: Where Immersive Wave Experiments Make a Difference

Immersive wave experimentation turns a small, reflective laboratory into a controlled environment where physical and virtual wavefields can be combined. This is useful when the desired experiment is too large, too low-frequency, or too difficult to build directly. A few examples are:

• Mixing Real and Virtual Wave Scattering. Prototype complex samples by combining a tangible physical core with a flexible virtual surround. Measure once, then remix shapes, impedances, and arrangements digitally—no need for re-machining after every tweak.
• Virtually Created Media and Metamaterials. Build phononic crystals and metamaterials in software; let your lab “feel” them in real time. Explore phenomena like gain media or parity–time (PT) symmetric responses that would be impossible to achieve with traditional passive materials.
• Independent Car Audio. Imagine creating multiple immersive zones—one per passenger—by surrounding each listener with compact arrays that couple to their own virtual acoustic space. The result? A personalized soundscape delivered without overwhelming the cabin.
• VR-Room Acoustics for Learning and Play. Step into “impossible” rooms that morph from the echo of a cathedral to the silence of an anechoic box. Students and audiences can walk through these dynamic environments while the same wavefield is measured in real time.
• Hybrid Metamaterials and Phononic Crystals. Embrace the best of both worlds: keep one physical unit cell while virtually tiling the surrounding lattice. This fusion lets you experiment with sweeping changes in geometry, bias, and non‑physical parameters (like gain, loss, or asymmetry) between measurements—without having to rebuild your setup.
• Nonlinear, Time‑Varying Media & PT Symmetry. Inject controlled nonlinearity or time modulation directly into your numerical models, then tether them to the real wavefield. Mix a virtual gain medium with a physical lossy element to craft direction‑dependent transmission and absorption effects that passive samples simply can’t sustain.
• Digital Twins and Cloning. Capture a scatterer’s Green’s functions and replay them holographically. In doing so, you can “clone” the object on demand—rotating it, moving it, or digitally modifying its impedance at the drop of a hat.
• Low‑Frequency, Broadband Rigs and Collocated Apertures. Future-ready modular 3D arrays can span entire rooms (≈2.7 × 3.1 × 4.5 m) and dive deeper into lower frequency regimes without needing bulky absorbers. By collocating emit/record surfaces, we can suppress higher‑order coupling, making first‑order Green’s components the stars of the show—thanks to low‑latency control.
• Open Recipes for Safety and Reproducibility. The idea is to eventually share kernels, calibration sets, and level‑safe presets so that others can reproduce—and responsibly deploy—immersive experiments. In this way, the lab isn’t just editing waves; it’s editing the rules of the game.

Beyond these immediate applications, immersive wave experimentation opens doors to entirely new realms of physics and wave control strategies. By blending physical and virtual worlds, our lab becomes more than a measuring instrument—it evolves into a dynamic canvas on which we sculpt the flow of energy and matter.

There’s also a deeper scientific edge to this work. Elastic immersive wave experimentation unlocks a low‑frequency window (≈1–20 kHz in solids) that conventional small labs struggle to explore. In this regime, wavelengths are comparable to sample sizes, providing an ideal testbed for homogenization theory, scale‑bridging ideas, and advanced techniques like time reversal and virtual source methods. This is where the lab not only pushes the envelope of wave manipulation but also pioneers potential breakthroughs in seismic imaging and non‑destructive testing.

Graph adapted from: Elastic Immersive Wave Experimentation (ETH Zurich, 2022).

Overall, immersive experiments provide a practical way to connect measured wavefields with virtual environments and to test wave-control ideas under laboratory conditions.

References