Qudit error-correction models #
Konno et al. demonstrated a propagating optical GKP qubit, while Brock et al. demonstrated error-corrected GKP qutrits and ququarts beyond break-even. These records extract the finite algebra and performance bookkeeping needed to reason about a higher-dimensional OAM design. They do not claim that either experiment implemented an OAM qudit or that a 100-dimensional code has been constructed.
Addition of finite cyclic qudit labels, with the modulus positivity made
explicit for the Fin result.
Instances For
A zero shift leaves a finite cyclic label unchanged.
The binary symplectic phase pairing generalized to a finite cyclic qudit
module. It is a ring expression over ZMod dimension; no field assumption is
made, so composite dimensions such as 100 remain admissible.
Instances For
The cyclic qudit phase pairing is antisymmetric.
GKP qudit geometry extracted from the displacement-operator construction.
- dimension_at_least_two : 2 ≤ dimension
- logicalPauliSpacing : ℝ
- logicalPauliSpacing_nonnegative : 0 ≤ self.logicalPauliSpacing
- logicalPauliSpacingLaw : self.logicalPauliSpacing = √(Real.pi / ↑dimension)
- stabilizerLength : ℝ
- stabilizerLength_nonnegative : 0 ≤ self.stabilizerLength
- stabilizerLengthLaw : self.stabilizerLength = √(Real.pi * ↑dimension)
- envelopeParameter : ℝ
- envelopeParameter_pos : 0 < self.envelopeParameter
- meanPhotonNumber : ℝ
- meanPhotonNumber_nonnegative : 0 ≤ self.meanPhotonNumber
- finiteEnergyEnvelopeLaw : self.meanPhotonNumber = 1 / (2 * self.envelopeParameter ^ 2)
Instances For
The GKP qudit model exposes the logical displacement spacing.
The GKP qudit model exposes the stabilizer length.
A finite-energy GKP qudit QEC gain observation.
The source defines gain from physical and logical decay rates. A gain above one means the logical memory decays more slowly than the comparison physical qudit.
- dimension : ℕ
- dimension_at_least_two : 2 ≤ self.dimension
- physicalDecayRate : ℝ
- physicalDecayRate_pos : 0 < self.physicalDecayRate
- logicalDecayRate : ℝ
- logicalDecayRate_pos : 0 < self.logicalDecayRate
- gain : ℝ
- gain_nonnegative : 0 ≤ self.gain
- gainLaw : self.gain = self.physicalDecayRate / self.logicalDecayRate
- reportedStandardError : ℝ
- reportedStandardError_nonnegative : 0 ≤ self.reportedStandardError
- reportedGainAboveBreakEven : 1 < self.gain - self.reportedStandardError
- sourceStatus : String
Instances For
A physical/logical decay-rate comparison above break-even has gain above one.
The reported gain uncertainty still lies above the break-even point.
Centered integer labels for a 100-state OAM register.
Instances For
Centered OAM labels are distinct.
A 100-state OAM qudit design with explicit QEC-readiness fields.
The state is a finite OAM register. gkpModel records which higher-dimensional
GKP ideas are being compared; it is not an assertion that OAM modes realize the
oscillator displacement operators.
- state : OAM.Qudit 100
- modeLabel : Fin 100 → ℤ
- modeLabelLaw (mode : Fin 100) : self.modeLabel mode = centeredOAMMode mode
- modeLabel_injective : Function.Injective self.modeLabel
- modeSpacing : ℝ
- modeSpacing_pos : 0 < self.modeSpacing
- gkpModel : GKPQuditModel 100
- factorLeft : ℕ
- factorRight : ℕ
- factorizationLaw : self.factorLeft * self.factorRight = 100
- syndromeAxisCount : ℕ
- syndromeAxisCountLaw : self.syndromeAxisCount = 2
- syndromeReadoutSpecified : Bool
- recoveryMapSpecified : Bool
- modeLossRate : ℝ
- modeLossRate_nonnegative : 0 ≤ self.modeLossRate
- modeDephasingRate : ℝ
- modeDephasingRate_nonnegative : 0 ≤ self.modeDephasingRate
Instances For
The OAM design retains exactly 100 finite mode labels.
The two GKP stabilizer directions are represented as explicit design metadata, not as a completed OAM syndrome circuit.
The 100-level design can be factored as a 4-by-25 register at the finite arithmetic level. This does not assert an error-correcting tensor-product code.