Quantum Error Correction

Imperial College Team Reaches 50% Loss Threshold for Quantum Networks

Avatar of Ivy Delaney
October 2, 2026 by Ivy Delaney

Quantum computers can now reliably connect qubits across a network even when photon loss is unavoidable. Hybrid protocols have developed, combining established techniques with new methods utilising linear cluster states, to improve how information exchanges between matter-based quantum processing units. These advancements enable distributed logical operations and enhance qubit connectivity by increasing the effective distance over which reliable connections can be maintained; specifically, they demonstrate a fifty percent interface-erasure threshold in ideal conditions.

New methods exist for connecting individual quantum computers via networks, key for building machines capable of complex calculations. The focus was on improving how qubits, the fundamental units of quantum information, communicate across these connections, addressing unavoidable signal loss during transmission. By achieving a fifty percent reliability threshold alongside increased communication distances, an important step towards scalable distributed quantum computing where multiple processors work together effectively has been taken.

A sharp advance exists in connecting individual quantum computers into networks, vital for building machines capable of tackling problems beyond today’s technology. The challenge of maintaining reliable qubit communication across these connections despite unavoidable signal loss during transmission addressed; this can be imagined as ensuring walkie-talkies remain connected over increasing distances before their signals fade. These advancements allow multiple processors to work together effectively but require careful consideration of noise levels within each system; further details regarding protocol construction and performance outline below.

Hybrid protocols extend qubit connectivity via optimised syndrome extraction

Researchers and Rigetti Computing have increased interface distances for networked quantum processing units from d+1 to 2d+1. This improvement represents an advance in qubit connectivity, enabling reliable connections across greater network expanses than previously possible. Conventional Bell-pair interfaces sharply limited this range; however, these new techniques circumvent those restrictions through applying ZX calculus transformations, a method of optimising quantum circuits, to construct novel syndrome extraction methods within distributed rotated surface code lattice surgery.

These innovations facilitate loss-tolerant communication between QPUs. At low local error rates below 10-4, simulations revealed that the linear chain protocols exhibited more rapid improvements with decreasing erasure compared to conventional approaches, demonstrating their effectiveness at minimising noise impact.

Scaling quantum computers demands connection and this offers a pathway for linking matter-based processors via photons, particles of light, thus enabling collaborative computation beyond individual machine limits. Maintaining connectivity isn’t straightforward; current methods rely on fragile entanglement over links susceptible to signal loss during transmission, meaning reliable communication requires overcoming substantial hurdles in photon delivery. The new protocols achieve a fifty percent threshold for reliable communication despite potential errors, a key benchmark defining tolerable system noise levels. Current simulations account only for idealised resource states and do not yet demonstrate sustained performance under realistic manufacturing imperfections or complex environmental disturbances which remain significant challenges towards practical implementation. Combining established techniques with extended chains of entangled qubits, hybrid protocols substantially improve communication between networked quantum processing units and move beyond the limitations imposed by conventional connection strategies reliant solely on simple entanglement sharing.

The research demonstrated that several newly developed protocols achieved a fifty per cent interface-erasure threshold when local noise was absent, enabling more robust distributed operations between quantum processors. This matters because connecting individual quantum computers is essential to scaling up their computational power, but maintaining stable connections has proven difficult due to signal loss during transmission.

By utilising ZX calculus transformations and linear cluster states, these hybrid protocols extend the distance over which reliable communication can occur compared with previous methods using only Bell pairs. The authors calculated how performance diminishes with imperfect resource states and mapped correctable regions under various error types as part of this work.

👉 More information
🗞 Loss-tolerant distributed lattice surgery using fusion networks
✍️ Felix Burt, Richard Meister, Sheng-Ku Lin, Kuan-Cheng Chen, Michael Hanks, Roberto Bondesan, M. S. Kim and Kin K. Leung (Imperial College)
🧠 ArXiv: https://arxiv.org/abs/2610.01923

More like this
Quantum Error CorrectionResearchers Bound Gibbs-State Errors with Core-Tail Architecture
Quantum Error CorrectionResearchers Classify Complexity of Stabiliser State Problems
Quantum PhysicsMultiphoton control cuts time to build key quantum error correction codes
Quantum Error CorrectionResearchers Learn Quantum States from Fewest Possible Copies
Stay current

See today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals.

Tags:
quantum computers
Avatar of Ivy Delaney

Ivy Delaney

Ivy Delaney has been working with neural networks and machine learning since the mid-nineties, back when a couple of hidden layers and a long afternoon of training counted as ambitious. She has watched the field go from academic curiosity to the thing quietly running underneath everything, and she brings that long view to quantum computing. For Quantum Zeitgeist she covers the ground where the two fields meet. That means quantum machine learning and the variational algorithms it leans on, and it also means the less glamorous but more interesting story of classical machine learning already doing real work inside quantum machines, decoding error-correcting codes, calibrating noisy hardware and learning the error models that simulators depend on. She writes about the hardware those algorithms have to run on too, and about the post-quantum cryptography scramble that the same hardware has set off. Her stories typically start with the paper, whether that is peer-reviewed work, conference proceedings or an arXiv preprint, with the source linked so you can hold a claim up against the research it came from. She is unimpressed by benchmarks that will not say what they beat, and by demonstrations that only work in the press release.

Latest Posts by Ivy Delaney: