Limits for recovering information from uncertain noise

A new method uses polynomial projections to find the optimal recovery from coherent rotation.

Recovery from a coherent rotation with an unobserved angle depends on the encoding and calibration. This recovery depends on what calibration reveals about the noise.

Two polynomial projections organize the answer for a known finite-dimensional generator. A physical Krylov chain separates correctable error amplitudes from their first state-dependent distortion.

Projection onto lower-angle powers identifies the part that a coherent decoder can cancel. Under explicit rank and moment conditions, this gives the exact leading optimal entanglement infidelity at weak noise.

The optimal worst-input error lies between one and dd times that infidelity for a logical dimension dd. When the angle distribution is unknown, one decoder is optimized for every distribution compatible with the calibration.

If the sole calibration datum is βr=E[sin⁡2r(θ/2)]\beta_r=\mathbb{E}[\sin^{2r}(\theta/2)], the optimal worst-prior infidelity is O(βr)O(\beta_r) when generator powers through degree r−1r-1 are correctable.

In a loss-free four-photon memory, population measurements at two calibration durations retain the fully informed quartic error coefficient. A single duration permits loss quadratic in the noise width.

Key facts

  • The optimal worst-input error is between one and dd times the entanglement infidelity for logical dimension dd.
  • A physical Krylov chain separates correctable error amplitudes from their first state-dependent distortion.
  • Population measurements at two calibration durations in a four-photon memory retain the fully informed quartic error coefficient.

All sections

Edition