Practical blueprint for low-depth photonic quantum computing with quantum dots Ming Lai Chan† ,1, 2, ∗ Aliki Anna Capatos,3, 4, † Peter Lodahl,1, 2 Anders Søndberg Sørensen,2 and Stefano Paesani2, 4, ‡ 1
Sparrow Quantum, Blegdamsvej 104A, DK-2100 Copenhagen Ø, Denmark Center for Hybrid Quantum Networks (Hy-Q), The Niels Bohr Institute, University of Copenhagen, DK-2100 Copenhagen Ø, Denmark 3 Quantum Engineering Centre for Doctoral Training, University of Bristol, Bristol, United Kingdom 4 NNF Quantum Computing Programme, Niels Bohr Institute, University of Copenhagen, Blegdamsvej 17, DK-2100 Copenhagen Ø, Denmark (Dated: July 23, 2025)
arXiv:2507.16152v1 [quant-ph] 22 Jul 2025
2
Fusion-based quantum computing is an attractive model for fault-tolerant computation based on photonics requiring only finite-sized entangled resource states followed by linear-optics operations and photon measurements. Large-scale implementations have so far been limited due to the access only to probabilistic photon sources, vulnerability to photon loss, and the need for massive multiplexing. Deterministic photon sources offer an alternative and resource-efficient route. By synergistically integrating deterministic photon emission, adaptive repeat-until-success fusions, and an optimised architectural design, we propose a complete blueprint for a photonic quantum computer using quantum dots and linear optics. It features time-bin qubit encoding, reconfigurable entangledphoton sources, and a fusion-based architecture with low optical connectivity, significantly reducing the required optical depth per photon and resource overheads. We present in detail the hardware required for resource-state generation and fusion networking, experimental pulse sequences, and exact resource estimates for preparing a logical qubit. We estimate that one logical clock cycle of error correction can be executed within microseconds, which scales linearly with the code distance. We also simulate error thresholds for fault-tolerance by accounting for a full catalogue of intrinsic error sources found in real-world quantum dot devices. Our work establishes a practical blueprint for a low-optical-depth, emitter-based fault-tolerant photonic quantum computer. I.
INTRODUCTION
Building a useful quantum computer rests on designing a practical framework that integrates platform constraints, error correction, and scalable control systems. A blueprint for such a framework must bridge theory and hardware capabilities and describe precise physical component requirements, their integration and overhead, and performance targets under realistic conditions that may guide experimental progress. Currently, blueprints tailored to different physical platforms [1–9] have been proposed, addressing a range of aspects including specific hardware design, physical error modelling and resource estimation. Among the candidate platforms, photonic quantum computing offers fast gate times, scalability, modularity, and room-temperature operation, but faces three stumbling blocks. First, resource-state generation based on spontaneous parametric down conversion is highly inefficient due to the low success probability [10]. Since each physical qubit in a resource state is usually encoded in multiple photons, which requires additional probabilistic gates, the overall success rate is very low. Preparing a logical qubit of code distance d = 20 is estimated to require 2d3 = 16, 000 such states [11]. Alternatively, continuous-variable systems use bright beams
∗ Ming-Lai.Chan@sparrowquantum.com † These authors contributed equally to this work. ‡ stefano.paesani@nbi.ku.dk
that enable deterministic entanglement generation [12], yet these schemes are highly sensitive to loss and require non-Gaussian photon operations that are typically implemented probabilistically by photon subtraction. Second, current photonic architectures typically rely heavily on complex routing involving fibre delays and multiplexing. This introduces large optical depths in resourcestate generation and/or photon measurement circuits, resulting in many loss channels which hinder scalability [2, 11, 13, 14]. Third, the current state of photonic hardware [15] places photon loss rates well above the thresholds required for fault-tolerant fusion-based quantum computation (FBQC) [11], and while existing methods can tolerate moderate photon loss (around 20% [16–25]), strategies for improving this tolerance without enormous overhead in the photon number per resource state [26, 27] have not been identified. To overcome these obstacles, recent photonic architectures leverage deterministic photon emitters to reliably produce high-fidelity entangled photonic resource states [24, 28–30], and adaptive repeat-untilsuccess (RUS) fusion gates to significantly boost tolerance to photon loss [31, 32]. Foundational experiments have already successfully demonstrated both ondemand resource-state generation [33–36] and fusion gates [21, 37], the essential building blocks for FBQC. With these theoretical and experimental advances, a critical next step is to integrate these techniques into a practical blueprint for emitter-based photonic architectures. However, several key aspects must be considered. First, current architectures primarily assume pathencoded photons, which may not align with the pho-