[Paper Review] Heisenberg-limited metrology with coherent control on the probes' configuration

Giulio Chiribella, Xiaobin Zhao|arXiv (Cornell University)|Jun 7, 2022

Quantum Information and Cryptography4 citations

TL;DR

This paper demonstrates that Heisenberg-limited metrology—achieving precision scaling as 1/N—can be achieved even under uncorrelated dephasing noise by using coherent superpositions of probe trajectories. By routing probes through multiple paths in superposition, the protocol restores Heisenberg scaling in both parallel (with respect to probe number/energy) and sequential (with respect to total time) configurations, overcoming a long-standing limitation in noisy quantum metrology.

ABSTRACT

A central feature of quantum metrology is the possibility of Heisenberg scaling, a quadratic improvement over the limits of classical statistics. This scaling, however, is notoriously fragile to noise. While for some noise types it can be restored through error correction, for other important types, such as dephasing, the Heisenberg scaling appears to be irremediably lost. Here we show that this limitation can sometimes be lifted if the experimenter has the ability to probe physical processes in a coherent superposition of alternative configurations. As a concrete example, we consider the problem of phase estimation in the presence of a random phase kick, which in normal conditions is known to prevent the Heisenberg scaling. We provide a parallel protocol that achieves Heisenberg scaling with respect to the probes' energy, as well as a sequential protocol that achieves Heisenberg scaling with respect to the total probing time. In addition, we show that Heisenberg scaling can also be achieved for frequency estimation in the presence of continuous-time dephasing noise, by combining the superposition of paths with fast control operations.

Motivation & Objective

• Overcome the fundamental limitation that uncorrelated dephasing noise typically destroys Heisenberg scaling in quantum metrology.
• Explore whether coherent control over probe configurations—specifically, superpositions of multiple trajectories—can restore Heisenberg scaling under noise.
• Demonstrate that trajectory superposition acts as a distinct quantum resource, independent of probe energy or total time.
• Extend the applicability of Heisenberg scaling to physically relevant noise models such as continuous-time Markovian dephasing.
• Provide both parallel and sequential protocols that achieve Heisenberg scaling under noise, using only single-particle superpositions and fast control.

Proposed method

• Implement a protocol where each probe is prepared in a coherent superposition of M alternative trajectories, each traversing an independent instance of the noisy process.
• Use interferometric measurements on the recombined paths to extract phase information, leveraging quantum coherence across configurations.
• Design a parallel protocol using N entangled probes, each routed through M = O(N) paths, to achieve Heisenberg scaling with respect to total energy.
• Develop a sequential protocol using a single probe that cycles through M = O(N) time steps, each involving a superposition of paths, achieving Heisenberg scaling with respect to total probing time.
• Apply fast control operations (e.g., path measurements and polarization shifts) at short intervals to simulate continuous evolution and enhance precision.
• Analyze the Fisher information for frequency estimation under the master equation model of continuous-time dephasing, showing asymptotic scaling F_ω ≥ T²/2 in the limit of vanishing time steps.

Experimental results

Research questions

• RQ1 Can Heisenberg scaling be restored in the presence of uncorrelated dephasing noise when probes are sent on definite trajectories?
• RQ2 Does coherent superposition of probe trajectories provide a new resource that enables Heisenberg scaling under noise, even when standard methods fail?
• RQ3 Can both parallel and sequential protocols achieve Heisenberg scaling using trajectory superposition, and how does the scaling depend on probe energy and total time?
• RQ4 What is the precision limit for frequency estimation under continuous-time Markovian dephasing when fast control and path superposition are combined?
• RQ5 How does the superposition of trajectories compare to other quantum resources like entanglement or total energy in enhancing metrological precision?

Key findings

• Heisenberg scaling is restored in the presence of uncorrelated dephasing noise when probes are routed through a coherent superposition of M = O(N) paths, both in parallel and sequential protocols.
• For the random phase kick model, the protocol achieves Heisenberg scaling with respect to the number of probes (or total energy) in the parallel setting.
• In the sequential setting, the protocol achieves Heisenberg scaling with respect to the total probing time T, with Fisher information scaling as F_ω ≥ T²/2 in the limit of infinitesimal time steps.
• The superposition of trajectories enables Heisenberg scaling even for continuous-time Markovian dephasing, a noise model previously thought to prevent such scaling.
• The protocol’s precision is independent of the number of paths M when M grows as T/t, showing that trajectory superposition acts as a distinct quantum resource beyond energy or time.
• The results are robust under fast control operations and remain valid even when the path degree of freedom is subject to imperfections, suggesting feasibility for experimental realization with current photonic technologies.