🔬 applied physics

Tailoring multiple scattering acoustic media with perfect transmission for non-Abelian braiding and beyond

This paper proposes and experimentally demonstrates a multi-scattering-based approach using generalized Wigner-Smith operators to design reflectionless acoustic media with unitary transmission matrices, enabling perfect transmission that performs arbitrary unitary operations, including non-Abelian braiding of waveguide modes.

Original authors: Hongkuan Zhang, Guancong Ma

Published 2026-07-14

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Hongkuan Zhang, Guancong Ma

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

Imagine sound waves as a chaotic crowd of people trying to rush through a crowded hallway. Usually, if you throw a bunch of obstacles (like metal cylinders) into that hallway, the crowd gets messy. They bump into each other, bounce off walls, and get scattered in every direction. Most of the energy gets stuck or reflected back, and the original order of the crowd is lost forever. This is what scientists call "multiple scattering," and for a long time, it was seen as a problem that scrambled signals and ruined control.

But in this new study, researchers Hongkuan Zhang and Guancong Ma at Hong Kong Baptist University turned that chaos into a superpower. They asked a bold question: What if we could arrange those obstacles so perfectly that the sound passes through without losing a single drop of energy, but instead, it rearranges the "strands" of the sound waves in a specific, magical way?

The Magic Trick: Reflectionless Complex Media
The team designed a special "complex medium"—a section of an acoustic waveguide (a tube for sound) filled with 80 tiny aluminum cylinders. Think of these cylinders not as obstacles, but as a highly organized dance floor. By using a smart computer algorithm based on something called "generalized Wigner-Smith operators" (a fancy tool that calculates how to push the cylinders just right), they found the perfect positions for every single cylinder.

When they did this, something amazing happened. The sound didn't just pass through; it became "reflectionless." This means none of the sound bounced back. Instead, the entire wavefront was perfectly transmitted, but it had been transformed. The researchers showed that they could make the sound waves perform a "unitary operation," which is a fancy math way of saying they could swap, mix, or rotate the different modes (or "lanes") of the sound wave with 100% efficiency.

The Braiding Dance
The most exciting part of their discovery is something called "non-Abelian braiding." Imagine four ribbons of sound flowing through the tube. In the old days, to make these ribbons braid around each other (like hair being plaited), scientists had to use very long, delicate structures and move things very slowly, like a slow-motion dance. It was fragile and hard to do.

Zhang and Ma showed that by stacking their special "chaotic" sections, they could braid the sound ribbons instantly. They created three different "generators" (let's call them Move A, Move B, and Move C).

Move A swaps the first and second ribbons.
Move B swaps the second and third.
Move C swaps the third and fourth.

Here is the mind-bending part: The order matters. If you do Move A then Move B, the ribbons end up in a different place than if you do Move B then Move A. This is called "non-Abelian" behavior. In their experiments, they proved this by stacking the sections. When they swapped the order, the sound waves came out completely different. This is a huge deal because it means disorder (the random-looking cylinders) can be used to build logic gates for sound, similar to how computers use logic gates for electricity.

What They Did and Didn't Do
The researchers didn't just guess this would work; they built it. They set up a 2D waveguide that was 20 cm wide and 2 meters long. They operated at a frequency of 3.2 kHz, which allowed four different "modes" of sound to travel through. They placed 80 cylinders, each with a radius of 2 cm, into a 35 cm long section.

They explicitly ruled out the idea that you need slow, smooth changes (adiabatic evolution) to braid waves. Their method is fast and relies on the "messy" scattering of the cylinders. They also showed that this isn't just about swapping ribbons; they could make the sound perform other tasks, like a "NOT gate" (flipping a signal) or even a "Fast Fourier Transform" (a mathematical tool used in signal processing), though the Fourier Transform part was demonstrated through simulation, not a physical experiment yet.

How Sure Are They?
The team is very confident in their physical results. They measured the sound waves with microphones and found that the "reflection" (the sound bouncing back) was almost zero, and the "transmission" (the sound getting through) was perfect, matching their computer models almost exactly. They tested this with 40 different random starting positions for the cylinders, and every time, the optimization worked.

However, they are careful to note a few limits. Their design works best at one specific frequency (3.2 kHz). If you change the pitch even a little bit, the magic starts to fade, but they found it stays very good (over 90% effective) for a range of 160 Hz around that note. They also noted that real-world air absorbs a tiny bit of sound, but this loss is so small and predictable that they can mathematically "clean it up" to see the perfect result.

The Big Picture
This work suggests that we don't need perfect, clean, orderly structures to control waves. Sometimes, a carefully designed mess is better. By treating disorder as a resource, they created a compact platform (just 35 cm long, or about 3.2 wavelengths) that can do complex operations on sound waves. This could lead to new ways of sending multiple messages at once (multiplexed communication), better imaging, or even manipulating quantum waves in the future. But for now, they have simply shown that with the right arrangement of cylinders, sound can dance in ways we never thought possible.

Technical Summary: Tailoring Reflectionless Complex Media for Non-Abelian Braiding of Acoustic Modes

Problem Statement
While multiple scattering in complex media has been harnessed for diverse wave control applications—such as time-reversal refocusing, wavefront shaping, and deep energy delivery—the full control of the transmission matrix ( $T$ -matrix) remains a significant challenge. In generic complex media, the scattering matrix ( $S$ -matrix) typically possesses limited open transmission channels, resulting in low energy transmission coefficients where much of the incident energy is reflected or absorbed. Furthermore, even in schemes achieving perfect transmission (e.g., via complementary complex media), the outgoing wavefronts are often scrambled, offering little control over the specific waveform in each transmission channel. Consequently, realizing arbitrary, reflectionless unitary operations on wave modes within compact platforms has been difficult.

Methodology
The authors propose a multi-scattering-based approach to design reflectionless complex media capable of performing arbitrary unitary operations on incident wave modes. The core of their methodology involves:

System Design: The system consists of a two-dimensional (2D) rectangular acoustic waveguide containing a collection of identical metallic cylinders (scatterers). The waveguide supports four guiding modes at an operating frequency of 3.2 kHz.
Reflectionless Condition: The authors aim to design the medium such that the reflection matrices ( $R$ and $R'$ ) vanish ( $R = R' = 0$ ). Under this condition, the total $S$ -matrix becomes purely unitary and symmetric, and the total transmission matrix ( $T_{total}$ ) of cascaded sections follows simple matrix multiplication ( $T_{total} = T_K \cdots T_2 T_1$ ), enabling linear unitary operations on the input wave.
Optimization via Generalized Wigner-Smith Operators (GWSO): To achieve a target $S$ -matrix (specifically one with a desired $T$ -matrix and zero reflection), the authors employ a gradient-descent optimization algorithm. They treat the positions of $M=80$ cylindrical scatterers as variables on a smooth manifold. The optimization minimizes a cost function based on the squared Hilbert-Schmidt norm between the current $S$ -matrix and the target $S$ -matrix.
Gradient Calculation: The gradient of the cost function with respect to scatterer positions is calculated using the Generalized Wigner-Smith Operator (GWSO). The GWSO relates the change in the $S$ -matrix to the acoustic radiation force on the scatterers, allowing for the precise determination of how to move each scatterer to drive the system toward the target functionality.

Key Contributions and Results
The paper demonstrates the experimental realization of non-Abelian braiding of acoustic waveguide modes and the implementation of arbitrary unitary operations:

Non-Abelian Braiding: The authors experimentally realized the three generators ( $\sigma_1, \sigma_2, \sigma_3$ $σ_{1}, σ_{2}, σ_{3}$ ) of the braid group $B_4$ $B_{4}$ for four acoustic modes. By optimizing the scatterer configurations, they achieved media where the $T$ $T$ -matrices correspond to these braid generators, converting specific input modes to output modes with defined phase shifts (e.g., $\eta = \pi/6$ $η = π /6$ ).
- Experimental Verification: Measurements confirmed that the optimized media are reflectionless and perform the intended mode conversions. For instance, $\sigma_1$ converts the first and second modes while preserving the third and fourth, imparting specific phase delays.
- Non-Commutativity: By concatenating distinct scattering sections (e.g., $\sigma_1$ followed by $\sigma_2$ vs. $\sigma_2$ followed by $\sigma_1$ ), the authors demonstrated that the output modes depend on the order of operations, confirming the non-Abelian nature of the system. Conversely, commuting operations ( $\sigma_1$ and $\sigma_3$ ) yielded order-independent results.
Arbitrary Unitary Operations: Beyond braiding, the scheme was extended to implement other unitary operations, including bitwise NOT (BNOT) and controlled NOT (CNOT) gates, as well as a discrete Fourier transform (DFT). These were achieved by designing the $T$ -matrix to match the target logical or mathematical operations.
Robustness and Bandwidth: The study shows that the fidelity of the operation decays quadratically with frequency detuning and scatterer positioning errors. The optimized media maintain high fidelity (>0.9) over a bandwidth of 160 Hz around the 3.2 kHz center frequency and are robust against manual placement errors of up to 0.5 mm.

Significance and Claims
The authors claim that this work establishes complex media as a viable route for arbitrary modal control and operations on compact platforms, offering a new design paradigm for multimode, reconfigurable wave-based devices.

Departure from Adiabatic Methods: Unlike previous classical wave braiding realizations that rely on delicate adiabatic evolutions in slowly varying structures (requiring long propagation distances), this approach utilizes multiple scattering to synthesize unitary operations in much thinner structures (approximately 3.2 wavelengths thick).
Generalizability: The method allows for braiding with arbitrary phase factors ( $\eta$ ), overcoming limitations of previous methods restricted to $\eta = 0$ or $\pi$ . This capability opens the possibility of emulating anyonic exchange behavior in classical waves.
Scalability and Universality: The use of GWSOs provides a scalable and universal theoretical framework applicable to other wave types, including microwaves, elastic waves, and optics. The scheme does not require nonlinear effects, active gain, or intricate subwavelength resonances, making it less lossy and more broadband than metamaterial-based approaches.
Potential Applications: The authors suggest potential applications in multiplexed communication, imaging, and the manipulation of quantum waves, leveraging the unprecedented control over multiple-scattering waves demonstrated in this work.

More like this