Multiplexed Quantum Repeaters and Memory Requirements

Short answer: Multiplexing means giving each node of a quantum repeater chain many memory modes, so entanglement can be attempted many times in parallel instead of one try at a time. Proposals from 2007, 2021 and 2024 argue this can raise entanglement rates or relax memory requirements. These are theoretical analyses and proposals, with one key element of the 2024 work demonstrated in an experiment.

Why do quantum links need repeaters?

The 2007 Physical Review Letters paper by O. A. Collins, S. D. Jenkins, A. Kuzmich and T. A. B. Kennedy starts from the goal of long-distance quantum communication through distant pairs of entangled qubits, which it describes as a first step towards secure message transmission and distributed quantum computing. The authors say the most promising proposals need quantum repeaters to mitigate the exponential decrease in communication rate due to optical fibre losses, but that such repeaters are exquisitely sensitive to the lifetimes of their memory elements.

What did the 2007 proposal suggest?

The authors propose a multiplexing of quantum nodes that should enable quantum networks that are largely insensitive to the coherence times of the quantum memory elements. That is a proposal, stated in the abstract as something that should be possible, and the abstract reports no experiment.

How did the 2021 paper extend the idea?

A 2021 paper in New Journal of Physics notes that quantum repeater nodes based on quantum memories promise to overcome exponential signal decay, but that current memory performance hinders practical application. It says multimode solutions with multiplexing can offer a large increase in entanglement distribution rates. The authors propose a wavevector-multiplexed atomic quantum memory as the building block of a multiplexed repeater architecture. They report that it provides quasi-deterministic entanglement generation over extended distances, mitigating optical loss even with currently available memory devices, and that it exceeds repeaterless solutions and other repeater protocols such as temporal multiplexing. They also introduce the entangled-bit rate per number of employed nodes as a figure of merit for the cost-efficiency of larger inter-node distances. These performance statements are the authors' modelled results.

What does the 2024 preprint add?

The 2024 arXiv preprint, which has not necessarily been peer reviewed, studies chains whose nodes hold several memories, allowing multiplexing and entanglement distillation. The authors introduce quasi-local policies, in which nodes know more about the state of the chain than in fully local policies but not the full global state. They report improved performance over local policies, lower classical communication costs than global-knowledge policies, and better results than the well-known nested purification and doubling swapping policy in practically relevant parameter regimes. They also examine when entanglement distillation is useful and ask whether to distill before swapping or the reverse. Finally, they propose an experimental multiplexing repeater chain and report an experimental demonstration of a key element, a high-dimensional biphoton source, according to the visible abstract text, which is cut off at that point.

What to keep in mind

All three works are about how to organise repeater chains, and the performance claims come from the authors' models. The abstracts do not report a working multi-node repeater delivering entanglement over long distances. This post uses abstracts and visible summary text only, and full texts were not checked. It makes no claim about products or deployed networks.

FAQ

What is multiplexing in a quantum repeater?

Using multiple memory modes per node so that many entanglement attempts can run in parallel, which the 2021 abstract says can greatly increase entanglement distribution rates.

Does multiplexing remove the need for long memory lifetimes?

The 2007 authors propose that it should make networks largely insensitive to memory coherence times. That is a proposal.

What is a quasi-local policy?

A rule where each node uses knowledge of the connected region of the chain it belongs to, not the whole chain, according to the 2024 preprint.

Has a long-distance repeater been built from this?

Not according to these abstracts. They report theory, proposals and one demonstrated element.