Cat Code
| Bosonic quantum error correction using superpositions of coherent states (cat states) in cavity modes. Encodes qubit as |
0⟩ = ( |
α⟩+ |
-α⟩)/N and |
1⟩ = ( |
α⟩- |
-α⟩)/N. Protects against photon loss with exponential bit-flip suppression scaling as e^(-2α²). Developed by Mirrahimi, Leghtas, and Albert (2014). Experimentally implemented in superconducting circuits by Alice&Bob quantum computing. |
Properties
| Property |
Value |
| Category |
Error Correction |
| Sub-category |
Bosonic Quantum Code |
| Security status |
🧪 Experimental |
| Complexity |
Expert |
| Inventor |
Mazyar Mirrahimi, Zaki Leghtas, Victor Albert |
| Year |
2014 |
| Origin |
🇫🇷 France |
| Source |
algorithms/ecc/cat-code.js |
Security
Status: 🧪 Experimental
Known vulnerabilities
| Issue |
Description |
Mitigation |
| Phase-Flip Vulnerability |
Cat codes suppress bit-flip errors exponentially with |α|² but phase-flip errors increase linearly. Requires concatenation with outer codes (e.g., surface code) for full protection. Typical approach: cat code suppresses bit-flips, outer code corrects phase-flips. |
— |
| Coherent State Approximation |
Classical simulation uses truncated Fock basis representation. Real quantum implementation requires cavity QED hardware with strong dispersive coupling and multi-photon driven dissipation for autonomous error correction. |
— |
| Limited Distance |
Two-component cat code (S=1) can detect single photon loss. Higher-component codes (S>1) required for correcting multiple losses, increasing hardware complexity. |
— |
| Decoherence Time |
Cat state coherence requires cavity quality factor Q > 10⁶ and temperatures T < 50 mK. Experimental lifetimes reach 1-10ms for |α|=2, limiting gate operation speeds. |
— |
Documentation
References
Test vectors
9 vectors ship with this algorithm and run in the test suite. Byte values are hexadecimal.
| Vector 1 — [Encode logical |
0⟩ as even cat state with α=2.0](https://errorcorrectionzoo.org/c/two-legged-cat) |
| Field |
Value |
alpha |
2 |
input |
00 |
expected |
00 |
| Vector 2 — [Encode logical |
1⟩ as odd cat state with α=2.0](https://errorcorrectionzoo.org/c/two-legged-cat) |
| Field |
Value |
alpha |
2 |
input |
01 |
expected |
01 |
Vector 3 — Photon parity measurement on even cat state (no errors)
| Field |
Value |
alpha |
2 |
parityMeasurement |
Yes |
input |
00 |
expected |
00 |
Vector 4 — Photon parity measurement on odd cat state (no errors)
| Field |
Value |
alpha |
2 |
parityMeasurement |
Yes |
input |
01 |
expected |
01 |
Vector 5 — Encode with smaller coherent amplitude α=1.0
| Field |
Value |
alpha |
1 |
input |
00 |
expected |
00 |
Vector 6 — Encode with larger coherent amplitude α=3.0 (enhanced bit-flip suppression)
| Field |
Value |
alpha |
3 |
input |
01 |
expected |
01 |
Vector 7 — Single photon loss error detection and recovery
| Field |
Value |
alpha |
2 |
simulatePhotonLoss |
Yes |
input |
00 |
expected |
00 |
Vector 8 — Encode two logical qubits in separate cavity modes
| Field |
Value |
alpha |
2 |
input |
0001 |
expected |
0001 |
Vector 9 — Verify cat state fidelity remains high (F > 0.99) for α=2
| Field |
Value |
alpha |
2 |
checkFidelity |
Yes |
minFidelity |
0.99 |
input |
00 |
expected |
00 |
← All algorithms