⚛️ quantum physics

Hong-Ou-Mandel test to verify indistinguishability of the states emitted from a quantum key distribution transmitter implementing decoy Bennett-Brassard 1984 protocol

This paper presents and experimentally validates a practical Hong-Ou-Mandel interference test to verify the indistinguishability of pulses in a high-speed decoy BB84 quantum key distribution transmitter, demonstrating that modulation does not compromise security and providing a robust method for device certification without specific degree-of-freedom assumptions.

Original authors: Toshiya Tajima, Akihisa Tomita, Atsushi Okamoto

Published 2026-03-30

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Toshiya Tajima, Akihisa Tomita, Atsushi Okamoto

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: The "Perfect Twin" Test for Quantum Security

Imagine you are sending a secret message using light pulses. In the world of Quantum Key Distribution (QKD), these light pulses are like secret letters. To keep the message safe, the sender (Alice) encodes information into the "shape" or "timing" of the light.

The golden rule of this security system is: All the letters must look exactly the same to anyone watching, except for the secret code inside.

If an eavesdropper (Eve) can tell the difference between a letter encoded with a "0" and a letter encoded with a "1" just by looking at how the light behaves (its color, timing, or shape) outside of the secret code, she can steal the message without anyone knowing. This is called a "side-channel attack."

This paper is about a new, clever way to test if the light pulses are truly "indistinguishable" (identical twins) or if they have hidden "birthmarks" that give them away.

The Problem: The "Secret Handshake" Flaw

In a standard QKD system, the sender changes the state of the light to create different bits (0s and 1s).

The Ideal: Imagine two identical twins wearing different colored hats. If you only look at their faces, they are indistinguishable.
The Flaw: What if the twin wearing the "Red Hat" walks slightly faster, or has a slightly different voice, or smells like mint? Even though the hat (the secret code) is different, an observant spy could guess who is who just by the walk or the smell.

In quantum terms, the "walk" or "smell" is a physical property of the light pulse (like its exact arrival time or spectrum) that wasn't supposed to carry information but accidentally does.

The Solution: The "Hong-Ou-Mandel" (HOM) Dance

The researchers used a famous quantum physics trick called the Hong-Ou-Mandel (HOM) effect.

The Analogy: The Two-Dance Floor
Imagine a dance floor with a special mirror in the middle (a beam splitter).

You send two dancers (light pulses) toward the mirror from opposite sides.
If the dancers are perfect strangers (completely different), they might bump into each other, or one might go left and the other right. They act independently.
If the dancers are identical twins (indistinguishable), quantum magic happens. They refuse to split up. They always dance together, exiting the mirror on the same side.

The Test:

If you see the dancers exiting on opposite sides, they are different.
If they always exit on the same side (a "dip" in the number of times they split), they are identical.

The researchers built a machine to watch these "dancers" (light pulses) and see if they stick together.

What They Did in the Lab

They took a real-world QKD transmitter (the device that sends the secret keys) that uses a standard protocol called Decoy BB84. This device sends out pulses at a super-fast speed (1.25 billion times a second!).

They set up a test where:

They took two pulses right next to each other in time.
They made sure one pulse was in a "standard" state and the other was in a "modulated" state (the secret code).
They sent them into the HOM dance floor.

The Result:
They looked at the "dip" in the data.

The Dip: When the pulses are identical, the "splitting" events drop significantly.
The Finding: The dip was about 30% deep (visibility of 0.3). While not a perfect 100% (which is hard to get in the real world due to technical noise), the most important thing was that the dip looked exactly the same whether the pulses had secret codes or not.

Translation: The "Red Hat" twin and the "Blue Hat" twin walked at the exact same speed and had the exact same voice. The secret code didn't accidentally change their physical appearance. The system is safe from this specific type of spy.

Why This Matters (The "So What?")

No Assumptions Needed: Usually, to check if a device is safe, you have to assume, "We think the light is pure." This test doesn't need that. It just looks at the light and asks, "Are you identical?" It's a direct, physical proof.
Cheap and Easy: They used standard fiber-optic cables and off-the-shelf parts. You don't need a billion-dollar lab to do this check.
Future Proof: As QKD systems become more common in banks and governments, they need a standard "safety inspection." This paper proposes a standard test: Run the HOM test. If the pulses look the same, the device passes.

The "Imperfect" Reality Check

The researchers were honest about why the "dance" wasn't perfect (the visibility was only 0.3, not 1.0).

The Laser: The laser they used is like a slightly jittery drummer. It doesn't hit the beat exactly the same way every time.
The Detectors: The sensors are incredibly sensitive but have a tiny bit of "lag."
The Conclusion: Even with these imperfections, the difference between the coded and uncoded pulses was zero. The "jitter" was random noise, not a secret signal leaking out.

Summary in One Sentence

The authors invented a practical "twin test" using quantum interference to prove that a real-world quantum key sender doesn't accidentally leak secret information through the physical shape of its light pulses, ensuring the system is secure against sneaky eavesdroppers.

1. Problem Statement

Quantum Key Distribution (QKD) security relies on the assumption that transmitted pulses encoded with different modulation patterns (e.g., different BB84 states) are indistinguishable in all degrees of freedom (DOF) other than the encoded information.

The Vulnerability: If an eavesdropper can distinguish pulses based on non-encoded DOFs (such as spectrum, temporal waveform, or transverse mode) without altering the encoded bit, they can gain information without being detected (side-channel attacks).
Current Limitations: Traditional methods to verify indistinguishability involve measuring spectra or temporal waveforms using classical tools (optical spectrum analyzers, oscilloscopes). However, these methods are:
- Costly and time-consuming for single-photon levels.
- Potentially incomplete, as they may not exhaustively cover all non-encoded DOFs.
- Dependent on assumptions about specific physical properties.
The Goal: Develop a practical, quantum-optical method to verify pulse indistinguishability in high-speed QKD transmitters without relying on specific DOF assumptions.

2. Methodology

Theoretical Framework: SWAP Test vs. HOM Interference

The authors establish a theoretical equivalence between the SWAP test (a quantum algorithm for measuring state fidelity) and the Hong-Ou-Mandel (HOM) interference effect.

SWAP Test: Measures the overlap (fidelity) between two quantum states $|\phi\rangle$ and $|\psi\rangle$ . For mixed states (density matrices $\rho_1, \rho_2$ ), the outcome relates to the trace of their product, $\text{Tr}(\rho_1 \rho_2)$ .
HOM Equivalence: The authors demonstrate that for single-photon states, the HOM interference visibility is equivalent to the SWAP test outcome.
Extension to Weak Coherent Pulses (WCP): Since practical QKD systems use phase-randomized WCPs (not single photons), the authors derive the relationship between HOM visibility ( $V_{HOM}$ $V_{H O M}$ ) and the trace of the density matrix product.
- They define a "non-zero photon contribution" to fidelity to exclude vacuum state effects (which artificially inflate fidelity).
- Key Finding: $V_{HOM} \approx \frac{1}{2} \text{Tr}(\rho_a \rho_b)$ (for weak inputs). Thus, measuring HOM visibility directly characterizes the indistinguishability of the non-encoded DOFs.

Experimental Setup

Transmitter: A high-speed (1.25 GHz) QKD transmitter implementing the Decoy BB84 protocol using time-bin encoding.
- Source: Gain-switched laser diode (pulse width < 50 ps).
- Modulation: An asymmetric Mach-Zehnder interferometer (AMZI) creates double pulses. A dual-polarization IQ modulator encodes the pulses into BB84 states ( $X_0, X_1, Y_0, Y_1$ ).
- Pulse Thinning: An intensity modulator selects pulses at 51.2 ns intervals to avoid detector dead-time issues.
HOM Measurement:
- Interferometer: An asymmetric Michelson interferometer introduces a path difference ( $T'$ ) corresponding to one clock cycle (800 ps) plus a variable delay ( $\tau$ ).
- Detectors: Two Superconducting Nanowire Single-Photon Detectors (SNSPDs) with high efficiency and low jitter.
- Procedure: The system measures coincidence counts between adjacent pulses. One pulse is fixed in state $X_0$ , while the adjacent pulse is varied among $X_1, Y_0, Y_1$ , and an unmodulated reference.
- Scan: The path difference $\tau$ is scanned around 0 ps to observe the HOM dip.

3. Key Contributions

Theoretical Equivalence: Proved that HOM visibility for phase-randomized WCPs is a direct measure of the trace of the product of density matrices ( $\text{Tr}(\rho_1 \rho_2)$ ), effectively serving as a SWAP test for indistinguishability without vacuum bias.
Practical Protocol: Developed a standardized, fiber-optic based test using off-the-shelf components and SNSPDs to certify QKD transmitters.
High-Speed Validation: Successfully applied the test to a 1.25 GHz transmitter, a speed regime where previous indistinguishability tests were difficult to perform.
Statistical Rigor: Introduced a rigorous statistical analysis (Likelihood Ratio Test and ANOVA) to determine if modulation affects indistinguishability, rather than just observing raw visibility values.

4. Results

HOM Visibility: The observed HOM visibility was approximately 0.3 across all state combinations ( $X_1-X_0, Y_0-X_0, Y_1-X_0$ $X_{1} - X_{0}, Y_{0} - X_{0}, Y_{1} - X_{0}$ ) and the unmodulated reference.
- Note: The value is lower than the theoretical maximum of 0.5 (for pure states) due to mixed-state effects (spectral/temporal jitter) inherent in gain-switched lasers, but this is consistent with literature.
Statistical Significance:
- Visibility Differences: The differences in visibility between modulated and unmodulated states were less than 0.035.
- Hypothesis Testing: A Likelihood Ratio Test yielded a p-value of 0.18, and ANOVA on dip positions yielded p = 0.2.
- Conclusion: The null hypothesis (that all states are drawn from the same population/indistinguishable) cannot be rejected at the 5% significance level.
Dip Characteristics: The HOM dip width (7–8 ps FWHM) was consistent across all modulation patterns, indicating that modulation did not introduce significant timing jitter or spectral distortion.

5. Significance and Implications

Security Certification: This method provides a device-independent (regarding specific DOFs) way to verify the "indistinguishability assumption" critical for QKD security proofs. It confirms that the modulation process in the tested transmitter does not leak information via side channels.
Standardization: The approach uses standard fiber-optic components and single-photon detectors, making it a viable candidate for industry standards in QKD system evaluation and certification.
Mixed State Handling: The paper clarifies that while mixed states (common in practical lasers) reduce visibility, they do not invalidate the test as long as the degree of mixing is independent of the modulation pattern.
Future Application: The method is adaptable to other encoding schemes (e.g., polarization) by adding appropriate analyzers, offering a universal tool for QKD implementation security.

In summary, the paper successfully demonstrates that HOM interference is a robust, practical, and theoretically sound metric for certifying the security of high-speed QKD transmitters against side-channel attacks arising from pulse distinguishability.