🔬 optics

Full-Field Mode Sorter for Optical Knots

This paper demonstrates a proof-of-principle full-field sorter using optimized phase-only elements to distinguish topologically structured optical knots, such as Hopf links and trefoils, with high accuracy and low crosstalk, thereby enabling practical readout for knot-based high-dimensional optical communication.

Original authors: Tareq Jaouni, Roohollah Ghobadi, Ebrahim Karimi

Published 2026-06-23

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Tareq Jaouni, Roohollah Ghobadi, Ebrahim Karimi

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 Idea: Untangling Light "Knots"

Imagine you have a bundle of laser beams. Most lasers are like straight, boring pencils. But scientists have figured out how to twist these beams into complex, 3D shapes that look like actual knots—like a trefoil (a simple three-loop knot), a cinquefoil (a five-loop knot), or a Hopf link (two interlocked rings).

These "optical knots" are exciting because they could carry a lot more information than a standard laser beam. Think of it like this: a standard laser is like a single letter of the alphabet (A, B, or C). An optical knot is like a whole word or a complex symbol. If you can send messages using these knots, you can pack way more data into a single beam of light.

The Problem:
Sending the message is easy. But how do you read it? If you shine a knot-shaped beam at a detector, the machine doesn't know if it's looking at a "trefoil" or a "cinquefoil." It's like trying to identify a specific type of tangled headphone cord just by looking at a blurry pile of wire. You need a way to untangle them and sort them into different boxes so you know exactly which one you have.

The Solution: The "Smart Light Sorter"

The authors of this paper built a device called a Full-Field Mode Sorter. Think of this device as a highly intelligent traffic cop for light.

The Setup: They use special glass-like screens called phase masks. You can imagine these as invisible, programmable filters that don't block the light but change its shape slightly, like a lens that bends light in very specific, weird ways.
The Training: They didn't just guess how to make these filters. They used a computer program (a "Genetic Algorithm," which works like evolution) to "breed" the perfect filter designs. The computer tried thousands of designs, kept the ones that worked best, and tweaked them until they were perfect.
The Result: When a knotted beam hits this sorter, the filter reshapes the light so that:
- If it's a Trefoil, it gets pushed to the Left box.
- If it's a Hopf Link, it gets pushed to the Right box.
- If it's a Cinquefoil, it gets pushed to the Center box.

How Well Does It Work?

The researchers tested this with two different setups:

One Filter: Using just one special screen, they could sort two types of knots with about 90% accuracy. It was good, but not perfect.
Two Filters: By stacking two of these screens, the accuracy jumped to over 95%. It was much better at telling the knots apart, even when they looked very similar.

They also tested if the sorter would break if the light wasn't perfect. In the real world, light might be slightly tilted, shifted to the side, or have ripples in it (like looking at a reflection in a wavy pool).

The Test: They rotated the knots, moved them slightly off-center, and added "ripples" (aberrations) to the light.
The Verdict: The sorter was surprisingly tough. Even when the input light was messy or slightly wrong, the sorter could still tell the knots apart most of the time.

Why This Matters (According to the Paper)

The paper claims this is a major step forward because:

It's a New Way to Read Light: Unlike other methods that rely on simple numbers or codes, this method looks at the entire complex shape of the light field.
It Handles "Messy" Data: Optical knots are naturally "messy" (mathematically speaking, they aren't perfectly distinct from each other). This sorter is designed specifically to handle that messiness and still find the differences.
It's Ready for the Real World: Because it works even when the light is slightly distorted, it could be used for secure communication systems where the signal might get jumbled up during travel.

The Bottom Line

The authors have created a "magic lens" system that can take a beam of light twisted into a complex knot, look at it, and say, "Ah, that's a Trefoil knot!" with high accuracy. They proved that using two of these lenses works better than one, and that the system is robust enough to handle real-world imperfections. This opens the door to using these fancy 3D light shapes to send more information faster and more securely.

Technical Summary: Full-Field Mode Sorter for Optical Knots

Problem Statement
Optical knots—topologically structured light fields where phase or polarization singularities trace linked or knotted trajectories—represent a promising high-dimensional alphabet for optical information and quantum communication. However, their practical application is hindered by the lack of a robust readout method. Unlike standard spatial modes (e.g., Orbital Angular Momentum) which possess clear modal indices, optical knots are generally non-orthogonal and lack a single defining index. This non-orthogonality and the complexity of their field profiles make traditional mode-specific analytic sorters ineffective, leading to high crosstalk and degenerate mappings where distinct knots produce identical output patterns.

Methodology
The authors propose a proof-of-principle full-field mode sorter that utilizes one or two optimized phase-only elements (phase masks) to distinguish optical knots. The approach adapts the architecture of full-field sorters previously used for general transverse modes but tailors the optimization specifically for the non-orthogonal nature of knots.

Optical Alphabet: The study focuses on an alphabet of three specific knots: the Hopf link, the trefoil, and the cinquefoil. These are generated using Gaussian-weighted Milnor polynomials, where shape parameters control the knot geometry.
Sorter Architecture: The device employs a linear optical measurement scheme. It uses either a single phase plane followed by a lens ( $U_1$ ) or two phase planes separated by propagation and lenses ( $U_2$ ). The phase masks act as multiplicative operators $\Phi_j(r) = \exp(i\phi_j(r))$ .
Optimization Strategy: The phase masks are optimized using a Genetic Algorithm (GA). The objective function is a scalar fitness metric ( $F$ $F$ ) derived directly from the output intensity distributions, balancing three factors:
1. Balanced Sorting Contrast ( $C_b$ ): Rewards correct assignment while penalizing leakage, with specific weighting for the worst-performing channel to ensure uniform performance across the alphabet.
2. Communication-Motivated Error Penalty ( $R_d$ ): Based on the average symbol error rate, representing the secret-key rate potential for high-dimensional states.
3. Determinant-Based Distinguishability ( $V_P$ ): A factor $|\det(p)|$ that penalizes solutions where different input knots map to linearly dependent output distributions, thereby avoiding degenerate mappings.
Simulation and Testing: Numerical simulations (780 nm wavelength, 128x128 grid) were conducted to generate the phase masks. The sorter's performance was benchmarked against ideal conditions and various experimental imperfections, including input rotations, transverse translations, and phase aberrations (modeled via Zernike polynomials). A real-world underwater channel simulation was also performed using average Zernike coefficients.

Key Results

Sorting Performance:
- Single-Plane Sorter: Successfully sorted two-knot combinations (e.g., cinquefoil and Hopf link) with efficiencies between 18–28% and sorting probabilities ( $P$ ) exceeding 90%. Performance was inversely correlated with the modal overlap of the specific knot pair. Sorting all three knots simultaneously with a single plane yielded lower efficiency (7%) and sorting ability (80%), highlighting fundamental limitations for non-orthogonal modes.
- Two-Plane Sorter: Utilizing two optimized phase planes significantly improved performance. For two-knot combinations, efficiencies reached 25–35% with $P > 95\%$ . For the full three-knot alphabet, the system achieved an efficiency of 13% and a sorting ability of 93%, demonstrating that high distinguishability can be maintained even as dimensionality increases.
Robustness:
- Rotations: The sorter exhibited modulation in crosstalk dependent on the rotation angle, with modulation frequencies corresponding to the topological properties of the knots (e.g., fivefold for cinquefoil).
- Translations: Performance declined as the input beam moved away from the center, though the decline was not radially symmetric and varied by knot type.
- Aberrations: The sorter remained robust under phase aberrations (e.g., Zernike modes) provided the aberration aperture was not too small relative to the beam waist. However, decreasing the aperture size or increasing aberration strength caused significant drops in efficiency and sorting ability.
- Real-World Channel: In a simulated underwater channel, the sorter maintained low crosstalk (e.g., 0.48% for the 3-knot alphabet) and average efficiencies around 13.8%, with errors falling below standard security thresholds.

Significance and Claims
The paper claims to provide the first full-field sorter specifically designed for topologically structured light. Its primary significance lies in:

Extending Sorting Capabilities: It successfully extends full-field optical mode sorting to optical knots, a class of fields previously difficult to sort due to non-orthogonality and the lack of modal indices.
Practical Readout: It offers a viable readout route for knot-based high-dimensional optical communication, demonstrating that distinct knots can be distinguished with low crosstalk using minimal phase elements.
Empirical Guidelines: The work provides an empirical prescription for knot shaping to achieve optimal theoretical distinguishability.
Limitations and Future Directions: The authors modestly note that the current method relies on field information at a single transverse plane rather than topological invariants. They suggest that future improvements could involve auxiliary output channels to capture inconclusive contributions or the use of linear diffractive neural networks, but they do not claim these are currently implemented.

The study concludes that while sorting efficiency decreases with higher dimensionality, the sorting ability (distinguishability) remains robust, making the proposed sorter a viable component for secure, high-dimensional quantum communication systems involving optical knots.