Figure

Description

A photonic qubit architecture based on single-photon encoding (typically dual-rail or polarization) with state preparation, gates, and measurement implemented using linear optical elements, single-photon sources, and photodetectors.

A common encoding is dual-rail, where the logical states are defined by a single photon occupying one of two spatial modes:

Beam splitters and phase shifters implement single-qubit unitaries as passive transformations that conserve photon number. Entangling gates are induced through measurement and feed-forward (KLM-style), trading deterministic interactions for optical network + ancilla overhead.

The foundational result of Knill, Laflamme, and Milburn (KLM, 2001) established that scalable universal quantum computation is possible using only linear optics, single-photon sources, and photon detectors — with no photon-photon nonlinearity required. The key insight is that measurement-induced nonlinearities, combined with teleportation-based gate constructions and offline resource state preparation, can achieve asymptotically deterministic entangling gates at the cost of significant ancilla overhead.

Hamiltonian

In integrated linear optics, passive transformations are generated by a quadratic bosonic Hamiltonian:

which induces a unitary mode transformation acting on creation operators as:

Single-qubit operations in dual-rail encoding are realized with beam splitters and phase shifters (Mach-Zehnder interferometers), while effective entangling operations are implemented by measurement-induced nonlinearities (ancilla photons + feed-forward in KLM-style protocols).

Motivation

  • Photons have negligible decoherence during transmission, making this architecture naturally network-compatible
  • The KLM result established that scalable quantum computing is possible with linear optics + measurement alone — no photon-photon interaction needed
  • Provides a central route to distributed and communication-first quantum architectures
  • Room-temperature operation for the photonic components (only detectors require cryogenics)

Experimental Status

KLM foundational result — Knill, Laflamme, and Milburn (2001):

  • Proved that efficient quantum computation is possible using only linear optics, single-photon sources, and photon detectors
  • Probabilistic nonlinear sign (NS) gate succeeds with probability 1/4; CZ gate with probability 1/16 in the basic scheme
  • Teleportation-based constructions boost effective success probability at the cost of ancilla overhead

Experimental demonstrations:

  • Two-photon entangling gates demonstrated using linear optics and post-selection (O’Brien et al., 2003)
  • Integrated photonic circuits on silicon and silicon nitride chips have realized high-visibility single-qubit operations
  • Boson sampling demonstrations validated the linear optical platform

Scaling challenges:

  • Loss and source/detector efficiency remain the dominant practical scaling limits
  • Photon indistinguishability (Hong-Ou-Mandel interference) is critical for gate operation
  • KLM overhead motivates alternative approaches (cluster-state MBQC, fusion-based QC)

Key Metrics

MetricValueNotesFidelity reference
Native 2Q interactionMeasurement-induced (probabilistic)No direct photon-photon interactionKnill et al. 2001
Deterministic universal QC pathYes (with ancilla + feed-forward)Core KLM result; large overheadKnill et al. 2001
Hardware bottleneckLoss + source/detector efficiencyDominant practical scaling limitKok et al. 2007
Operating temperature300 K (optics) / 1–4 K (detectors)SNSPDs require cryogenics

References

Original proposal

  • E. Knill, R. Laflamme, and G. J. Milburn, “A scheme for efficient quantum computation with linear optics,” Nature 409, 46 (2001)

Reviews

Linked Papers