[Paper Review] Metrology with entangled coherent states - a quantum scaling paradox

Michael J. W. Hall|arXiv (Cornell University)|Jul 10, 2013

Quantum Mechanics and Applications2 references3 citations

TL;DR

This paper resolves a quantum scaling paradox in phase metrology by showing that entangled coherent states (ECS) cannot achieve better than $1/n^{1/4}$ average phase resolution, despite quantum Cramer-Rao bounds suggesting $1/n$ scaling. The limitation arises from vacuum background noise and bias in estimation, which prevent optimal performance even with nonlinear schemes, and highlights the need for iterative, multicomponent probe strategies to approach theoretical bounds.

ABSTRACT

There has been much interest in developing phase estimation schemes which beat the so-called Heisenberg limit, i.e., for which the phase resolution scales better than 1/n, where n is a measure of resources such as the average photon number or the number of atomic qubits. In particular, a number of nonlinear schemes have been proposed for which the resolution appears to scale as 1/n^k or even exp(-n), based on optimising the quantum Cramer-Rao bound. Such schemes include the use of entangled coherent states. However, it may be shown that the average root mean square errors of the proposed schemes (averaged over any prior distribution of phase shifts), cannot beat the Heisenberg limit, and that simple estimation schemes based on entangled coherent states cannot scale better than 1/n^{1/4}. This paradox is related to the role of 'bias' in Cramer-Rao bounds, and is only partially ameliorated via iterative implementations of the proposed schemes. The results are based on new information-theoretic bounds for the average information gain and error of any phase estimation scheme, and generalise to estimates of shifts generated by any operator having discrete eigenvalues.

Motivation & Objective

• To resolve the apparent contradiction between quantum Cramer-Rao bounds predicting $1/n$ scaling for entangled coherent states and actual achievable phase resolution.
• To clarify the operational limitations of the quantum Cramer-Rao bound in phase estimation, particularly its dependence on local unbiasedness.
• To establish general information-theoretic bounds on average phase estimation error and information gain that are valid regardless of prior knowledge or probe state structure.
• To compare the performance limits of entangled coherent states with unentangled coherent states and NOON states in terms of average resolution and information gain.
• To investigate whether iterative or multicomponent probe schemes can recover the promised scaling of the Cramer-Rao bound for ECS, especially in the presence of vacuum contributions.

Proposed method

• Derives new information-theoretic bounds for the average root mean square error (RMSE) and mutual information in phase estimation, based on the entropy of the generator operator and its asymmetry.
• Applies these bounds to probe states with discrete eigenvalue generators, including phase shifts generated by photon number operators.
• Compares the new bounds with the quantum Cramer-Rao bound for entangled coherent states and factorisable coherent states across varying average photon numbers $n$.
• Analyzes the canonical phase distribution of entangled coherent states, showing a uniform background component due to vacuum contributions that degrade phase resolution.
• Evaluates the feasibility of iterative phase estimation schemes using multicomponent probe states to recover $1/n$ scaling, despite the presence of vacuum noise in ECS.
• Uses numerical and analytical methods to compare scaling behaviors for small and large $n$, focusing on the asymptotic performance of ECS versus unentangled coherent states.

Experimental results

Research questions

• RQ1 Can entangled coherent states achieve phase resolution better than $1/n^{1/4}$ in the average case, as suggested by the quantum Cramer-Rao bound?
• RQ2 Why do quantum Cramer-Rao bounds overestimate the achievable phase resolution for entangled coherent states, particularly in the presence of vacuum components?
• RQ3 To what extent can iterative or multicomponent probe schemes recover the $1/n$ scaling promised by the Cramer-Rao bound for ECS?
• RQ4 How does the vacuum background in entangled coherent states affect the canonical phase distribution and the resulting phase estimation error?
• RQ5 What is the fundamental limit on average information gain and phase resolution for any phase estimation scheme, independent of probe state structure?

Key findings

• The average root mean square error (RMSE) for phase estimation using entangled coherent states cannot scale better than $n^{-1/4}$, even though the quantum Cramer-Rao bound suggests $1/n$ scaling.
• The vacuum component in entangled coherent states introduces a uniform background in the canonical phase distribution, which degrades phase resolution and limits the effectiveness of standard estimation schemes.
• Unentangled coherent states achieve a $1/n^{1/2}$ scaling in average phase resolution, outperforming entangled coherent states asymptotically despite the latter's higher Cramer-Rao bound prediction.
• The quantum Cramer-Rao bound has limited operational significance because it only applies to locally unbiased estimators, and the region of unbiasedness is typically comparable in size to the promised resolution.
• Iterative implementations using multicomponent probe states may recover the Cramer-Rao promise, but require significantly larger scaling constants and are not straightforwardly applicable to entangled coherent states due to vacuum noise.
• Information-theoretic bounds based on generator entropy or asymmetry provide more robust and general limits on phase estimation performance than the Cramer-Rao bound, especially in the presence of prior uncertainty or nonlinearity.