Building a quantum internet

A quantum internet is a quantum network that uses processors at different geographical locations. Our goal is to develop this technology, and enable quantum communication between any two places on earth. One important application of a quantum internet is secure communication, with privacy guaranteed by the fundamental laws of physics.


Quantum processors can be connected into a quantum network, creating a large quantum computing cluster. This approach is called networked quantum computing and it offers a natural path towards scalability, complementing quantum computing work described in the previous chapter. Combining a quantum internet and a networked quantum computer allows remote users/providers to perform secure quantum computing ‘in the cloud’.

Quantum network protocols and applications

Long-term goal: Development of a full control stack and novel applications for a quantum internet.

Highlights:

  • We designed the first networked operating system for quantum devices, QNodeOS. An implementation on a microcontroller will allow testing on quantum hardware in 2021. A novel quantum network layer protocol (ACM) has been implemented, and a patent application is in process. As part of this effort, we have also designed an instruction set for quantum network nodes, NetQASM, formalized and integrated with QNodeOS through an SDK and a compiler.

  • To control large scale networks, a time-division multiple access network architecture for quantum networks was developed, for which a patent application has also been initiated. In order to actually realize such architectures, a software-defined network architecture based on programmable quantum data planes has been developed (ACM). To find requirements of future quantum network architectures, we have developed a method for parameter optimization using machine learning in collaboration with SurfSARA. (arXiv)

  • This year has seen the public release of NetSquid, a unique quantum network simulator capable of exploring and validating all aspects of future quantum network architectures – ranging from the physical layer, over the control plane, to an understanding of application requirements. As an example of its application, we have simulated entanglement distribution over a 1000-node repeater chain in <2 seconds, and a control-plane of a quantum switch beyond the analytically known regime (arXiv).

  • We obtained efficiently computable boundaries for the fundamental limits at which entanglement can be distributed in a quantum network. This limit can be used to benchmark implementations of the first quantum networks (Physics). Furthermore, we developed algorithms for computing the rates at which realistic protocols can deliver entanglement, and for the quality of the entanglement produced. These algorithms can evaluate protocols for repeater chains with thousands of segments, and are exponentially faster than previous state-of-the-art algorithms (IEEE Journal).

Quantum network experiments

Long-term goal: Experimental demonstrations, some of which are already compatible with existing telecom infrastructure, that push the frontier of quantum networking and yield new techniques and insight.

Highlights:

  • We have developed new techniques to probe and control the charge state of the NV center. Using these techniques, we have unravelled the orbital and spin dynamics of NV centers, enabling single-shot spin readout of the neutral charge state and measurement of second-long relaxation times. These findings and methods are important for using NV centers in remote entanglement-based networks (Phys. Rev. Lett.).
  • We further continued the development of a quantum memory for light using cryogenically-cooled thulium-doped crystals. Our recent findings include the demonstration of optical coherence times in excess of 1 millisecond – the second-longest time reported to date for any rare-earth crystal – as well as storage of pulses of light, in optical coherence, over 100 microseconds. Furthermore, we have identified a promising non-ground-state transition at telecom wavelength that has never before been investigated for quantum applications.

Quantum processors for quantum networks

Long-term goal: To develop and exploit multi-qubit quantum processors for use in a quantum internet and networked computing, with a current focus on diamond-based spin qubits.

Highlights:

  • We have developed a machine learning method to characterise qubit couplings efficiently and automatically in NV center based devices. We showed that we could rapidly identify the interactions of the NV center with around ~30 qubits, using relatively simple experimental measurements. These results enable the fast, computerised and automatic characterisation of our quantum processors, which will be essential for realising larger-scale networks (arXiv).

  • We have realised initialisation, control and entanglement for a new type of qubit: the P1 center in diamond. This extends the number and types of qubits that can be controlled at a single NV center, and provides new opportunities for quantum processors, in particular the possibility of storing precious quantum states (data qubits) in such a way that they are isolated from the NV center dynamics while making entanglement over the network links (arXiv).