1. Introduction to MBQC

Measurement-based quantum computing (MBQC) is an alternative paradigm where computation proceeds through adaptive single-qubit measurements on a pre-entangled resource state, rather than sequential gate application.

Key Advantages

  • • Deterministic state preparation
  • • Feedforward-based error correction
  • • Natural fit for photonic systems
  • • Universal quantum computation

CV vs DV

  • • DV: Discrete qubits, photon counting
  • • CV: Continuous quadratures, homodyne
  • • CV enables deterministic gates
  • • Squeezed states as resources

2. CV Cluster States

A CV cluster state is a multimode entangled Gaussian state defined by a weighted graph G = (V, E). Each vertex represents a mode, and edges represent CZ (controlled-Z) interactions.

|G⟩ = Π(j,k)∈E CZjk(g) |p=0⟩⊗N

CZjk(g) = exp(ig x̂j x̂k)

Graph Topologies

TopologyModesApplication
Linear chainNSingle-qubit gates
Square latticeN×MUniversal QC, surface codes
Hexagonal6+1Topological protection
GHZ-rail2×NFault-tolerant encoding

3. Nullifier Formalism

Cluster states are uniquely characterized by nullifier operators—linear combinations of quadratures that have zero eigenvalue on the ideal state.

δ̂j = p̂j - Σk Ajk x̂k

Ideal: δ̂j |G⟩ = 0 for all j

Entanglement Witness

The nullifier variance ⟨δj²⟩ serves as an entanglement witness:

  • • ⟨δ²⟩ < 1 SNU → Genuine multipartite entanglement
  • • ⟨δ²⟩ = e-2r/2 for infinite squeezing → 0
  • • Practical systems: ⟨δ²⟩ ≈ 0.2-0.5 SNU with 6-10 dB squeezing

4. Covariance Matrix Formalism

For Gaussian states, all information is contained in the first moments (displacement) and second moments (covariance matrix).

σij = ½⟨{Δξ̂i, Δξ̂j}⟩

ξ = (x₁, p₁, x₂, p₂, ..., xN, pN)ᵀ

Squeezed Vacuum

σ = ⊕ⱼ diag(e⁻²ʳ/2, e²ʳ/2)

Nullifier Variance

⟨δⱼ²⟩ = cᵀ σ c

5. Symplectic Transformations

Gaussian unitary operations correspond to symplectic transformations on the covariance matrix:

σ' = S σ Sᵀ

S Ω Sᵀ = Ω (symplectic condition)

Key Operations

Phase shifter: S(φ) = R(φ) ⊕ R(φ) (rotation by angle φ)
Beam splitter: S(θ) mixes modes with cos(θ), sin(θ)
CZ gate: Adds correlations: p → p + g·x

6. Interferometer Mesh Architectures

Clements Decomposition

  • • Symmetric arrangement
  • • ⌈N/2⌉ layers depth
  • • Internal phase shifters
  • • Balanced loss distribution
  • • N(N-1)/2 beam splitters

Reck Decomposition

  • • Triangular/staircase
  • • N-1 layers depth
  • • External phase shifters
  • • Simpler construction
  • • N(N-1)/2 beam splitters

7. Loss Modeling

Optical loss is modeled as a beamsplitter interaction with a vacuum mode:

σlossy = η σideal + (1-η) σvac

η = 10-LdB/10 per component

Critical Threshold

For 8-mode square graphs with 6 dB squeezing: ~0.7 dB/layer preserves entanglement (⟨δ²⟩ < 1 SNU)

8. Phase Optimization

The compiler solves a constrained optimization to find mesh phases that minimize nullifier variance:

minimize J({φ,θ}) = (1/N) Σj ⟨δj²⟩

subject to mesh realizability constraints

References

1. Menicucci, N. C. et al. "Universal quantum computation with continuous-variable cluster states" PRL 97, 110501 (2006)

2. Weedbrook, C. et al. "Gaussian quantum information" Rev. Mod. Phys. 84, 621 (2012)

3. Clements, W. R. et al. "Optimal design for universal multiport interferometers" Optica 3, 1460 (2016)

4. Reck, M. et al. "Experimental realization of any discrete unitary operator" PRL 73, 58 (1994)

5. Yokoyama, S. et al. "Ultra-large-scale continuous-variable cluster states" Nature Photon 7, 982 (2013)