⚛️ quantum physics

Multipartite device-independent quantum key distribution using W states

This paper demonstrates that multipartite device-independent quantum key distribution is feasible using W states by constructing suitable Bell inequalities and proposing a long-distance protocol that outperforms existing GHZ-based schemes in terms of transmission distance and detection efficiency requirements.

Original authors: Makoto Ishihara, Wojciech Roga, Masahiro Takeoka

Published 2026-03-03

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Makoto Ishihara, Wojciech Roga, Masahiro Takeoka

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: A New Way to Lock Secrets Together

Imagine you and a group of friends want to create a secret code that only you all know. You want to do this even if you don't trust the lockboxes (the devices) you are using to make the code. This is called Device-Independent Quantum Key Distribution (DI-QKD).

Usually, scientists have used a specific type of "quantum glue" called a GHZ state to tie everyone together. Think of a GHZ state like a perfectly synchronized dance troupe. If one dancer stops, the whole group stops. It's very powerful, but it's also very fragile. If one dancer trips (loses a photon), the whole dance falls apart. This makes it hard to send these codes over long distances, like across a city or between countries.

This paper asks a bold question: "Can we use a different kind of quantum glue, called a W state, to do this job?"

The answer is YES. And not only can we do it, but the W state might actually be better for long-distance secrets.

The Characters: GHZ vs. W States

To understand the difference, let's use an analogy of magic coins.

The GHZ State (The "All-or-Nothing" Coin): Imagine you have three magic coins. If you flip them, they are all Heads, or they are all Tails. But if you lose one coin, the magic is gone. The connection breaks completely. This is great for short distances, but if you try to send these coins through a long, leaky pipe (a fiber optic cable), you likely won't get any coins through at all.
The W State (The "Resilient" Coin): Now imagine a different set of magic coins. If you flip them, only one of them is Heads, and the others are Tails. But here's the magic: if you lose one coin, the remaining two are still connected! They know that one of them is the "Heads" coin, even if they don't know which one. The W state is robust. It can survive losing a few pieces of itself.

The Problem: The W state is harder to use for making a secret code because the "Heads" coin is random. You can't just say, "We all have Heads," because you don't know who has it. You have to do a lot of extra math (error correction) to figure out the pattern.

The Solution: The authors of this paper figured out a new way to play the game. They invented a new set of rules (called Bell Inequalities) that are specifically designed to work with the W state's "resilient" nature.

How They Did It: The Three Steps

1. Inventing the New Rules (The Bell Inequalities)

In the quantum world, to prove you have a secret key, you have to prove that your devices are behaving in a way that is impossible for a spy to fake. This is done by breaking a "rule" called a Bell Inequality.

The authors used a computer to design a custom-made rulebook specifically for W states. They found a way to measure the W state that creates a huge "violation" of the rules. This huge violation acts like a loud alarm that says, "Hey! No spy is here! We are truly connected!" This alarm is loud enough to overcome the extra math work needed to fix the W state's randomness.

2. Checking the Hardware (Detection Efficiency)

Every time you try to catch a quantum particle, your detector might miss it (like trying to catch a fly with a net that has holes).

The team calculated that to use W states, your "net" needs to be very good (about 97% to 99% efficient).
While this is hard for light-based systems, it's possible with matter-based systems (like trapped atoms).
They showed that even with these high requirements, it is mathematically possible to generate a secret key.

3. The Long-Distance Trick (The RIHT Protocol)

This is the most exciting part. How do you send these W states over 100 kilometers of fiber optic cable without them getting lost?

The Old Way (GHZ): You generate the whole group of entangled particles in one place and try to send them all out. If the cable is long, the signal dies. It's like trying to shout a message across a canyon; the wind (loss) drowns it out.
The New Way (W State + RIHT Protocol): Instead of sending the whole group, everyone sends a small piece of their own entangled state to a central station in the middle.
- Imagine everyone throws a ball to a central referee.
- The referee catches the balls and performs a "magic interference" trick (single-photon interference).
- If the referee sees exactly one ball light up, it means the group is connected!
- Because everyone only sent a small piece, the signal doesn't die out as fast. It's like everyone whispering a piece of a secret to the middle, and the middle person puts the puzzle together.

The Result: The authors showed that this new method can distribute secret keys over 100+ kilometers, whereas the old GHZ method struggles to go more than a few kilometers.

Why This Matters

Think of the current quantum internet as a network of short, fragile bridges. You can cross them, but they break easily if the water gets rough (signal loss).

This paper proposes building floating rafts (W states) instead. They are a bit wobbly and require more people to balance them (more error correction), but they can float over much rougher, longer waters.

In summary:

We found a new type of quantum glue (W states) that is tougher against loss than the old type.
We invented new rules to prove this glue is secure against spies.
We showed a new way to send it over long distances, potentially unlocking the ability to have secure, device-independent secret keys between cities, not just between neighbors.

This opens a new path toward a global, unhackable quantum internet.

1. Problem Statement

Quantum Key Distribution (QKD) allows remote parties to share secret keys with information-theoretic security. While standard QKD relies on trusted devices, Device-Independent QKD (DI-QKD) guarantees security based solely on the violation of Bell inequalities, removing the need to trust the internal workings of the devices.

Existing multipartite DI-QKD protocols (often called conference key agreement) primarily rely on Greenberger-Horne-Zeilinger (GHZ) states. However, GHZ states are fragile; they suffer from exponential loss rates when distributed over long distances in lossy channels, and they are difficult to generate and maintain.

W states are an alternative form of multipartite entanglement known for being more robust against particle loss and easier to distribute over long distances. However, a major open question remained: Can W states be used for multipartite DI-QKD?

The Challenge: Unlike GHZ states, where measuring in the $X$ -basis yields perfectly correlated bits, measuring a W state in any basis results in inherent errors (correlations are not perfect). This necessitates a non-zero error correction cost, which reduces the key rate. If the security gain (Eve's uncertainty) does not sufficiently exceed this error correction cost, no secure key can be distilled.

2. Methodology

The authors propose a framework to overcome the inherent imperfections of W states in DI-QKD through three main methodological steps:

A. Construction of Tailored Bell Inequalities

To maximize the security (Eve's uncertainty) relative to the error correction cost, the authors developed a numerical optimization approach:

They formulated the problem of finding a Bell inequality that maximizes the conditional von Neumann entropy $H(A|x^*, E)$ (Eve's uncertainty about Alice's key) while minimizing the error correction cost.
Using the Navascués-Pironio-Acín (NPA) hierarchy, they relaxed the optimization of the guessing probability into a Semidefinite Programming (SDP) problem.
By solving the dual problem of the SDP, they extracted explicit Bell expressions (linear combinations of probabilities) specifically tailored to be largely violated by W states in various scenarios (e.g., 3, 4, and 5 parties).

B. Analysis of Detection Efficiency Requirements

The authors analyzed the minimum detection efficiency ( $\eta_e$ ) required to generate a positive key rate under realistic conditions:

Scenario 1: Arbitrary Pauli measurements (theoretical limit).
Scenario 2: Displacement-based measurements followed by photon detection (practical optical implementation).
They utilized a numerical optimization based on quasi-relative entropy to calculate key rates under imperfect detection, as the standard guessing probability approach is too sensitive to experimental imperfections.

C. Long-Distance Distribution via the RIHT Protocol

To address channel losses, the authors integrated their DI-QKD protocol with the Roga-Ikuta-Horikiri-Takeoka (RIHT) protocol:

Instead of generating a W state at a central station and distributing it (where success probability scales as $O(\eta^N)$ ), each party locally prepares a two-mode entangled state and sends one mode to a central station.
The central station performs single-photon interference. A successful detection event (one photon detected among $N$ inputs) heralds the distribution of a W state.
This method changes the scaling of the distribution rate from exponential ( $O(\eta^N)$ ) to linear ( $O(\eta)$ ), similar to Twin-Field QKD.

D. Gaussian State Implementation

To verify experimental feasibility, they modeled the system using Gaussian states (Two-Mode Squeezed Vacuum states generated via Spontaneous Parametric Down-Conversion) and on-off detectors, rather than ideal qubits.

3. Key Contributions

Proof of Concept for W-State DI-QKD: The paper provides the first affirmative answer to whether W states can support multipartite DI-QKD. They demonstrate that by constructing specific Bell inequalities, the security gain can overcome the intrinsic error correction costs of W states.
Optimized Bell Inequalities: They numerically derived Bell inequalities for $(N, m, d)$ scenarios (specifically $N=3,4,5$ ) that are maximally violated by W states. For example, in the $(3,2,2)$ scenario, the conditional entropy is $\approx 0.998$ , while the error correction cost is $\approx 0.650$ , yielding a positive key rate.
Efficiency Thresholds:
- For arbitrary Pauli measurements, the required detection efficiency is 97.2% for 3 parties and 98.8% for 4 parties.
- For displacement-based measurements (more practical for optics), the requirements are higher (99.5% and 99.96%), highlighting the difficulty of fully optical implementations but suggesting feasibility with matter-based systems (e.g., trapped ions).
Long-Distance Protocol: By combining W states with the RIHT protocol, they show that secret keys can be distributed over >100 km. This significantly outperforms existing GHZ-based DI-QKD protocols, which are limited to a few kilometers due to exponential loss scaling.
Gaussian Feasibility: They demonstrated that the protocol remains viable even when implemented with experimentally feasible Gaussian states and on-off detectors, achieving longer distances than ideal GHZ-based protocols.

4. Key Results

Key Rates: The constructed protocols yield positive asymptotic key rates for 3, 4, and 5 parties.
- Table I in the paper shows that for the $(3,2,2)$ scenario, the key rate is 0.348.
Distance Comparison:
- GHZ-based (Direct Transmission): Limited to a few kilometers.
- W-based (RIHT Protocol): Achieves distances exceeding 100 km for 3 and 4 parties.
Parameter Sensitivity: The performance depends heavily on the parameter $q$ (the probability amplitude of the vacuum component in the local entangled states). A smaller $q$ increases fidelity but reduces the success probability; an optimal $q$ balances these to maximize the key rate over distance.
Bell Violation: The calculated quantum limits ( $L_q$ ) of the new Bell inequalities are almost identical to the values achieved by W states ( $L_W$ ), indicating that W states are near-optimal for these specific inequalities.

5. Significance

New Entanglement Resource: This work expands the toolkit for device-independent quantum information processing beyond GHZ states, establishing W states as a viable and potentially superior resource for multipartite networks.
Scalability: The use of the RIHT protocol with W states solves the "rate-distance" bottleneck that plagues GHZ-based networks. It offers a practical path toward long-distance quantum conference key agreement without quantum repeaters.
Experimental Roadmap: By analyzing both ideal Pauli measurements and practical displacement-based measurements (including Gaussian states), the paper provides a clear roadmap for experimentalists. It highlights that while fully optical setups require near-perfect detection, matter-based systems could achieve the necessary efficiencies sooner.
Fundamental Insight: The study deepens the understanding of the relationship between different classes of multipartite entanglement (GHZ vs. W) and Bell nonlocality, showing that W states, despite their lack of perfect correlations, can still support strong nonlocality sufficient for cryptographic security.

In conclusion, the paper successfully bridges the gap between theoretical multipartite entanglement and practical long-distance device-independent cryptography, proposing a robust protocol that leverages the unique resilience of W states.