Hacking is useless with quantum cryptography

For the first time, an international team led by LMU physicist Harald Weinfurter was able to use a more advanced form of cryptography. Encryption makes it harder for hackers to get in.

There is a lot of sensitive information on the Internet. Most of the time, sophisticated encryption methods make sure that this kind of content can't be intercepted and read. But in the future, high-powered quantum computers could break these keys in a matter of seconds.

So, it's a good thing that quantum mechanical techniques not only make it possible to make new, much faster algorithms, but also to make cryptography that works very well.

Quantum key distribution (QKD), as it is called in the jargon, is safe from attacks on the communication channel but not from attacks on or changes to the devices themselves.

So, the devices could send out a key that the manufacturer had saved and could have possibly given to a hacker. With DIQKD, which stands for device-independent QKD, it's a different story. In this case, the cryptographic protocol is the same no matter what device is being used.

Theoretically known since the 1990s, this method has now been put into practise for the first time by an international team of scientists led by LMU physicist Harald Weinfurter and National University of Singapore professor Charles Lim (NUS).

There are many different ways to exchange quantum mechanical keys.

Either light signals are sent from the sender to the receiver, or quantum systems that are entangled are used.

In this experiment, the physicists used two quantum mechanically entangled rubidium atoms that were in two laboratories on the LMU campus 400 metres apart from each other.

The two places are linked by a 700-meter-long fibre optic cable that goes under Geschwister Scholl Square in front of the main building.

To make an entanglement, scientists first send a laser pulse through each atom to make it move.

After this, the atoms fall back to their ground state on their own, and each gives off a photon.

Due to the conservation of angular momentum, the atom's spin is tied to the polarisation of the photon it sends out.

The two light particles travel through the fibre optic cable to a receiver station, where they are measured together. This shows that the atomic quantum memories are connected.

To exchange a key, Alice and Bob, which is what cryptographers usually call the two parties, measure the quantum states of their own atoms.

This is done randomly in two or four ways in each case.

If the directions are the same, entanglement makes the measurement results the same, so they can be used to make a secret key.

With the other measurements, a so-called Bell inequality can be calculated.

John Stewart Bell, a physicist, came up with these inequalities to see if nature could be explained with hidden variables.

The fact is that it can't, "says Weinfurter.

In DIQKD, the test is used to make sure that the devices haven't been tampered with. For example, it makes sure that hidden measurement results haven't been saved in the devices ahead of time "explains Weinfurter.

Experimental setup at LMU Campus

The implemented protocol, which was created by researchers at NUS, uses two measurement settings instead of one to generate keys. This is different from earlier methods, which only used one. By adding a second setting for key generation, it becomes harder to steal information, so the protocol can handle more noise and generate secret keys even for lower-quality entangled states "says Charles Lim.

With traditional QKD, on the other hand, security is only guaranteed if the quantum devices used have been well enough characterised.

So, people who use these protocols have to trust the specifications that QKD providers give them and hope that the device won't switch to a different mode of operation during the key distribution "Tim van Leent, who wrote the paper with Wei Zhang and Kai Redeker and was one of its four main authors, explains.

Since at least ten years ago, people have known that older QKD devices were easy to hack from the outside.

"With our method, we can now make secret keys with devices we don't know much about and that might not be reliable," says Weinfurter.

In fact, at first he wasn't sure if the experiment would work or not. But his team showed that his worries were unfounded and made the experiment much better, which he happily admits.

Along with the project that LMU and NUS worked on together, a group of researchers from the University of Oxford showed how device-independent key distribution works.

Researchers in the same lab used a system made up of two entangled ions to do this. These two projects lay the groundwork for future quantum networks that will let people in very different places talk to each other in a completely secure way "Charles Lim says so.

One of the next steps is to try to add more entangled atom pairs to the system.

This would make it possible to make a lot more entangled states, which would increase the data rate and, in the end, the security of the key "says van Leent.

The researchers would also like to make the range bigger. In the current setup, it was limited because the fibre between the laboratories lost about half of the photons.

In other tests, the researchers were able to change the wavelength of the photons into an area with low loss that could be used for communications.

So, with just a little bit of extra noise, they were able to make the quantum network connection work up to 33 kilometres away.