Bringing the quantum internet to life

An interview with Stephanie Wehner, Professor of Quantum Information at TU Delft

You are a thought leader on quantum networking. What first sparked your interest in quantum technologies?
I first learned about quantum technology a very long time ago, when I visited a public audience talk. I was immediately fascinated about the concept of quantum entanglement and the fact that qubits cannot be copied.

You are now a professor and the roadmap leader for Quantum Internet and Networked Computing at QuTech. What is your research focus?
My main research focus right now is two-fold. I’m working on defining a network stack, for which we’ve recently taken the first step by defining a link layer protocol. My other focus area is finding a scalable design, on how we can make a large-scale quantum internet a reality. For this, we built a simulation platform that can help us explore possible designs for a scalable quantum internet. This platform enables us to simulate, test and learn – for example – what hardware parameters are more important than others.

To make a quantum internet a reality, we must combine science and engineering. This is why I have many different disciplines working together in my team. This fits QuTech’s mission-driven culture, which means that we’re organised around roadmaps instead of departments. Our interdisciplinary way of working is also very much appreciated by our students. As an example, QuTech’s academy program for master and PhD students is open to all disciplines at the same time. Students from for example physics, engineering, and computer science work together on projects. They don’t necessarily always understand each other right away, but they appreciate that they know different things and complement each other.

What do you see as the key challenges in developing a quantum internet?
One key challenge is to generate entanglement faster, that is, we want to achieve faster data transfer over such a quantum internet to make it useful. At the same time, we need to reinvent networking from scratch. In most research areas, researchers have a more narrowly confined research scope. They are working on software for existing hardware, or developing new hardware that will work with known software protocols. In our case, we’re inventing the first version of quantum networks, all the way from application to hardware. This is massively complicated. And this is also why many parties are involved; it has never been done before, and all of the elements need to fit together.

What can we learn from the design of the classical internet?
We can learn a lot from the classical internet, for example how to prioritise and schedule data transmission and operations. We used this knowledge in the development of our link layer protocol. We can also learn from design mistakes that have been made. Of course, it is easy to say that in designing the quantum internet, we’re going to do everything right from the start. But in reality, we want to make the first small quantum network happen in three years. This means we’re making assumptions and pragmatic choices and it’s almost impossible to foresee the impact of the choices we’re making today. So even though we try to learn as much as possible from the classical internet, it is likely that we also won’t get it right the first time.

To what extent will the roles in the quantum internet be similar to the roles in the classical internet?
I think that the role division between people who develop and apply a quantum internet will be quite similar. This also makes it quite complex to develop a quantum internet, because we need many different players to make a quantum internet available and useable. I expect that many of the traditional players such as infrastructure and service providers or network equipment manufacturers in the classical internet will also move into the quantum internet, but there will also be new players. Today, it is not yet clear which of the existing players want to move into quantum networks, and what role they want to play. In addition, their success will partly depend on how fast they will act. New players and start-ups, although they often have less resources available, are more agile and can move faster.

You initiated the Quantum Internet Alliance, and are also the coordinator. What is the Quantum Internet Alliance working on?
The Quantum Internet Alliance is working towards the long-term goal of realising a quantum internet. We strive to develop the technology to send qubits over long distances, for which we need to develop quantum repeaters. Another goal is to advance functionality, meaning that we want to go beyond quantum key distribution and run more complicated applications on the network. We are developing a software and network stack that can do fast control, and that allows people to write arbitrary applications in software. This is not the case right now; at the moment everything is ad hoc. We also work on a Blueprint for a pan-European quantum internet that will tell us what to do next.

We will test our software stack on a four node network, by doing a remote quantum computation in the cloud. Although we will use a small quantum internet and quantum computers with only a few qubits, this will be a proof of concept for blind quantum computing. It will be truly secure quantum cloud computation, where nobody can look at the data nor the algorithm that is being used. In the future, this can have use cases like a manufacturer that wants to simulate a proprietary material design, but doesn’t want the service provider who owns the quantum computer to find out what this material design is.

How will you deal with intellectual property that will be developed by the partners of the Quantum Internet Alliance?
We have 23 partners from academia and industry, but not everyone is involved in everything. The intellectual property resides with the partner(s) who developed it. Partners who are not involved can gain access during the project if their own development depends on such access. It is not yet clear what will happen after this phase of the Quantum Internet Alliance ends, that is, whether there will be a continuation with the same partners continuing to use the same intellectual property during the project.

Will the quantum internet bring about revolutionary innovations?
I believe so, since many applications are already known and we do not even have a network yet. Entanglement, for example, can help in the coordination of tasks. I don’t know exactly how far you can push this, but I can imagine that there can be all kinds of new applications that will take advantage of this, even though they are difficult to envision today.

I believe that it is very important that already in this early stage of the development of a quantum internet, the technology is available to other people so they can experience how it works. If you look at the classical internet, then its success was largely derived by creative innovators who had access to the technology and who wrote programs to make this technology do fantastic new things like chatting remotely, sending pictures, or using Twitter. Today, the protocols and applications that people invented for a quantum internet are developed from pure theory. But this is not how successful applications usually come into existence. People need to be able to interact with a quantum internet hands-on, and play around with it to develop and try out new applications. I’m sure that the technology will spark off innovations, but we don’t know yet what they will be.

How does Europe compare to other regions when it comes to quantum networking?
In advanced quantum networking, Europe has a leading position. We have more quantum communication expertise than in, for example, the US. In Europe we have many different viewpoints and many different people, which is beneficial for technology development. However, to move fast and maintain our leadership position, better alignment between parties involved in research and development would be beneficial. I think we can learn from the space business, where there is much more alignment on the deliverables and goals. As an example, many parties are working on a single satellite. In quantum technology, everyone is developing their own. If our efforts are dispersed, it will be difficult to achieve ambitious ‘moonshot’ projects.

There are many fears about disruptive technologies. Should regulators get involved already?
We shouldn’t regulate things too early. If you regulate too early, you can confine the configuration space of technical development. Right now, we’re really just trying to make it work, and we can’t envision the entire impact, and the societal challenges it will bring. Nevertheless, it’s really important that different stakeholders start thinking about possible implications right now, and learn from innovation test beds as soon as possible.