Device-independent quantum key distribution at 100 kilometers achieved for the first time

In a future quantum communication network, a sufficiently large and reliable quantum computer (far beyond anything we currently have) could break the cryptographic keys used to protect our data and operations. The solution is to use new cryptographic keys based on the properties of quantum physics. These new keys are also vulnerable to certain types of attacks, but a particularly secure way to generate them is for the sender and receiver to share a quantum system with the celebrated property of quantum entanglement. Any attack would destroy the entanglement, which would be easily detectable by measuring the so-called Bell inequalities. This is the principle behind device-independent quantum key distribution (DI-QKD).

If we truly want DI-QKD to have practical applications in a future communication network, we must be able to perform it between network nodes separated by sufficiently large distances.

In this Science article, the researchers succeed in implementing DI-QKD between parties separated by distances ranging from 10 to 100 kilometers. The quantum systems that are entangled are neutral rubidium atoms.

The method used to entangle them was proposed, among others, by Ignacio Cirac in 2001 and consists of detecting the light emitted at a point halfway between the atoms. Since this light can originate from either atom indistinguishably, the rules of quantum physics tell us that, upon measuring a photon, the atoms become entangled. Although in this experiment the atoms are not actually separated by tens of kilometers—because they are located in the same laboratory—the distance is simulated by sending the light through coiled optical fiber cables of those lengths.

Quantum entanglement is a very fragile property: as light travels through the fiber, small losses accumulate, and the resulting entanglement is of lower quality, which translates into higher error rates in the generated cryptographic keys. The experimental results show that the key error rate ranges from 3% at a distance of 11 kilometers to more than 7% at 100 kilometers. Therefore, although this represents an important step in the right direction, we are still far from being able to achieve a completely secure, error-free quantum key distribution over intercity distances.

What does this advance consist of?

"Cryptography is a very important field today: every day, a large number of confidential information exchanges take place. Quantum key distribution (QKD) cryptographic protocols allow two honest users—usually referred to as Alice and Bob—to share a secret key, which can then be used to transmit confidential information privately. This represents a paradigm shift: in these schemes, the security of the protocol is guaranteed by the laws of quantum physics.

Around 2010, a series of successful attacks on QKD implementations showed that the security of these protocols can be compromised if the devices used in practice do not behave exactly as assumed in the theoretical description of the protocol. In other words, any mismatch between theory and experiment opens up opportunities for an adversary to break the security. To address this weakness, device-independent QKD protocols were introduced. In these protocols, the theoretical description makes no assumptions about the devices (it is independent of them), which are treated as black boxes. There is therefore no gap between theory and experiment that an adversary could exploit to break the communication, as in the earlier successful attacks. Device-independent protocols provide the highest level of security guaranteed by quantum physics. The quantum property used to achieve this exceptionally strong level of security is the violation of the so-called Bell inequalities, which is observed when measuring entangled particles.

The drawback of device-independent protocols is that their implementation is complex. Before this experiment, they had only been demonstrated in a configuration where the distance between Alice and Bob was two meters. The result was remarkable as a proof of principle, but in practice we do not usually need full cryptographic schemes to secure a transaction between two people who are only two meters apart. Taking a couple of steps would be more than enough 😊.

This article reports the experimental implementation of a device-independent protocol over distances of tens of kilometers, which are far more interesting from a practical point of view, and therefore opens the door to the viability of these protocols".

Is the study of good quality?

"Excellent. It is a major achievement and a very significant improvement over the state of the art: it is the first practical demonstration of device-independent quantum key distribution".

Are there any limitations that should be taken into account?

"Despite being an impressive advance, it still has aspects of a proof of principle. The most important one is that Alice and Bob are not separated by tens of kilometers. They are located in the same laboratory, but connected by a fiber whose length is tens of kilometers. In principle, this configuration simulates the situation in which Alice and Bob are located at sites separated by tens of kilometers and connected by a fiber, but in practice it is not exactly the same.

A slightly more technical point: the experiment reports that, for fiber distances of hundreds of kilometers, the conditions required to successfully carry out the protocol are observed, but a complete implementation has not yet been achieved".