Universal bounds for quantum metrology in the presence of correlated noise

TL;DR

This paper establishes fundamental quantum metrology bounds considering correlated noise, demonstrating how correlations affect precision limits and the potential advantages of entanglement and adaptive strategies.

Contribution

It introduces a general framework for deriving bounds in quantum metrology with correlated noise, improving upon existing bounds and analyzing their tightness and practical implications.

Findings

Negative correlations can enhance precision in certain dephasing scenarios

Bounds are tighter than previous results even for uncorrelated channels

Entanglement offers limited advantage in collisional thermometry

Abstract

We derive fundamental bounds for general quantum metrological models involving both temporal or spatial correlations (mathematically described by quantum combs), which may be effectively computed in the limit of a large number of probes or sensing channels involved. Although the bounds are not guaranteed to be tight in general, their tightness may be systematically increased by increasing numerical complexity of the procedure. Interestingly, this approach yields bounds tighter than the state of the art also for uncorrelated channels. We apply the bound to study the limits for the most general adaptive phase estimation models in the presence of temporally correlated dephasing. We consider dephasing both parallel (no Heisenberg scaling) and perpendicular (Heisenberg scaling possible) to the signal. In the former case our new bounds show that negative correlations are beneficial, for the latter we show evidence that the bounds are tight. We also apply the bounds to collisional thermometry, i.e. estimation of a parameter of the environment, showing evidence that entangled probes may provide only a limited advantage.