Quantum error correction is the set of techniques that protect quantum information against decoherence and operational noise by encoding a logical qubit redundantly across many physical qubits. Stabiliser measurements detect errors without collapsing the encoded state, allowing the system to diagnose and reverse bit-flip and phase-flip faults. It is the central prerequisite for fault-tolerant quantum computation, where logical error rates can be driven arbitrarily low provided physical error rates fall below a threshold.
- Quantum error correction protects fragile quantum information by encoding a logical Qubit redundantly across many physical qubits and measuring stabilisers to detect faults without destroying the encoded state. It generalises classical Error Correction to the constraints of quantum mechanics.
Overview
- Quantum states cannot be copied or measured directly without disturbance, so classical redundancy schemes do not transfer; quantum codes instead spread information non-locally and read out only error syndromes.
- A logical qubit is built from many noisy physical qubits, and repeated syndrome extraction identifies which correctable error occurred so it can be reversed.
- The threshold theorem shows that if physical error rates sit below a code-specific threshold, arbitrarily reliable Quantum Computation Paradigm computation is achievable by scaling the code distance.
Mechanisms
- Encode logical information across an entangled block of physical qubits.
- Measure commuting stabiliser operators to obtain a syndrome without collapsing the data.
- Decode the syndrome to infer the most likely error and apply a correction.
- Repeat continuously to combat ongoing decoherence and gate noise.
- Compose corrected operations into fault-tolerant logical Quantum Gate sequences.
Applications
- Realising fault-tolerant Quantum Computation Paradigm devices at scale.
- Preserving entanglement for long-distance Quantum Cryptography links.
- Underpinning resilient quantum Infrastructure and repeaters.
- Benchmarking hardware against the error-correction threshold.