Computing Library › Quantum Error Correction
Quantum Error Correction

Bosonic Codes Overview

Bosonic codes protect a qubit inside the many levels of a single oscillator, converting hardware-efficient encoding into a first line of error correction.

Hardware-efficient encoding

A single harmonic oscillator, a microwave cavity mode or an ion's motion, has infinitely many energy levels. Bosonic codes use this large space to store one logical qubit in a way that is robust against the oscillator's dominant error, single-photon loss. Because one physical component already provides redundancy, no array of qubits is needed for the innermost layer.

The main families

Kronos motion — error correction

Three families dominate. GKP codes use a grid of position and momentum and correct small displacements. Cat codes use superpositions of coherent states and engineer a strong noise bias. Binomial and other Fock-state codes use carefully weighted superpositions of photon-number states so that photon loss maps code words to detectable, correctable error words.

Bosonic codes shine as inner codes. A bosonic qubit with a nonzero distance can be the physical unit of an outer topological or LDPC code, so the outer code starts from an already-protected component. The break-even point, where an encoded bosonic qubit outlives its unencoded counterpart, has been demonstrated in superconducting cavities, a milestone the qubit-array codes reached later.

The limitation is control. Preparing non-classical oscillator states and performing gates on them requires an auxiliary nonlinearity, usually a transmon qubit, whose errors can propagate into the cavity. Managing that ancilla-induced error is the central experimental challenge of the bosonic approach.