arXiv · 2611.23083
Performance of Fusion-Based Quantum Computing with Imperfect Photon Indistinguishability
Abstract
Fusion-based quantum computation (FBQC) is a promising route to fault-tolerant photonic quantum computing, connecting small entangled resource states through linear-optical fusion measurements. While logical encoding can substantially improve resilience to photon loss, whether these benefits extend to other noise remain an open question. In this work, we show that this is not generally the case when accounting for partial photon distinguishability during resource state generation and fusion measurement. We develop hardware-motivated noise models to capture these effects and investigate their impact on fault tolerance in the 6-ring fusion network. We compare unencoded and Shor(2,2)-encoded resource states prepared using all-photonic and emitter-assisted repeat-until-success (RUS) schemes. For the all-photonic construction, encoding reduces the distinguishability threshold from 0.136% to 0.08%. In contrast, the RUS construction tolerates more distinguishability, with comparable unencoded and encoded thresholds (0.256% and 0.271%, respectively). These results expose a fundamental limitation of evaluating logical encodings under idealized resource-state assumptions and demonstrate that resource-state generation can critically determine the fault tolerance of photonic architectures.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Katia Hakem, Grégoire de Gliniasty, Nathan Shettell, Paul Hilaire, Stephen C. Wein. 2026-11-17. Performance of Fusion-Based Quantum Computing with Imperfect Photon Indistinguishability. https://arxiv.science/abs/2611.23083
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite arXiv Science by Indiciae: BibTeX · RIS.