Boosted Bell-state measurements for photonic quantum computation
Unlike superconducting or ion trap quantum computing, which uses entangling gates to perform the quantum computation, modern photonic quantum computing takes a different approach. Instead of entangling qubits as the computation runs, the idea is to prepare large entangled states ahead of time. These states are built up from smaller building blocks, starting with entangled pairs of photons.
A key challenge, however, is performing measurements that entangle pairs of photons, a process known as fusion. These measurements are crucial for fault-tolerant fusion-based quantum computing (FBQC), but when done with standard optical techniques, they only succeed about half the time. This low success rate limits the performance of FBQC systems.
Our recent work, published in Nature Portfolio’s Quantum Information, addresses this issue using a technique called boosting. In collaboration with Stefanie Barz’s group at the University of Stuttgart, we implemented a boosted fusion protocol that improves the fusion success rate by introducing extra resources.
In the experiment, we implemented a boosted measurement approach using a special optical device—a 4 x 4 multiport splitter—along with an additional entangled pair of photons. This setup allows the success rate to go up to 75%. In practice, we achieved a success rate of about 69.3%, which is a clear improvement over the standard 50%.
Beyond just improving the measurement success rate, the boosted approach also makes the overall quantum computing process more reliable. Specifically, it makes the system three times more resilient to photon loss, which is a common problem in these experiments, and it significantly reduces logical error rates.
These improvements are important steps toward making photonic quantum computing more practical and scalable. We are especially proud of how QC Design's Plaquette played a crucial role in simulating and understanding how fusion failures and other hardware imperfections impact quantum fault-tolerance performance.
This work was carried out as part of the PhotonQ project, supported by the Bundesministerium für Bildung und Forschung. We also acknowledge the contributions of the German Quantensysteme and IQST — Center for Integrated Quantum Science and Technology.
For full technical details, experimental methods, and results, the published paper is available in npj Quantum Information: https://www.nature.com/articles/s41534-025-00986-2.