Multiparameter estimation with single photons—linearly-optically generated quantum entanglement beats the shotnoise limit
Chenglong You, Sushovit Adhikari, Yuxi Chi, Margarite L. LaBorde, Corey T. Matyas, Chenyu Zhang, Zu-En Su, Tim Byrnes, Chao‐Yang Lu, Jonathan P. Dowling, Jonathan P. Olson
Abstract It was suggested in (Motes et al 2015 Phys. Rev. Lett. 114 170802) that optical networks with relatively inexpensive overheads—single photon Fock states, passive optical elements, and single photon detection—can show significant improvements over classical strategies for single-parameter estimation, when the number of modes in the network is small ( n < 7 ). A similar case was made in (Humphreys et al 2013 Phys. Rev. Lett. 111 070403) for multi-parameter estimation, where measurement is instead made using photon-number resolving detectors. In this paper, we analytically compute the quantum Cramér–Rao bound to show these networks can have a constant-factor quantum advantage in multi-parameter estimation for even large number of modes. Additionally, we provide a simplified measurement scheme using only single-photon (on–off) detectors that is capable of approximately obtaining this sensitivity for a small number of modes.
Abstract It was suggested in (Motes et al 2015 Phys. Rev. Lett. 114 170802) that optical networks with relatively inexpensive overheads—single photon Fock states, passive optical elements, and single photon detection—can show significant improvements over classical strategies for single-parameter estimation, when the number of modes in the network is small ( n < 7 ). A similar case was made in (Humphreys et al 2013 Phys. Rev. Lett. 111 070403) for multi-parameter estimation, where measurement is instead made using photon-number resolving detectors. In this paper, we analytically compute the quantum Cramér–Rao bound to show these networks can have a constant-factor quantum advantage in multi-parameter estimation for even large number of modes. Additionally, we provide a simplified measurement scheme using only single-photon (on–off) detectors that is capable of approximately obtaining this sensitivity for a small number of modes.
It was previously shown that optical networks using only single photon Fock states, passive linear optics, and single photon detection could achieve post-classical sensitivity for singleparameter estimation when the mode number is small. [1] Furthermore, it was shown that multiparameter estimation could achieve similar results with number-resolving detection. [2] Here, we consider an analogous architecture to the Quantum Fourier Transform Interferometer(QuFTI) proposed in [1] for the estimation of multiple phases simultaneously. Thus, we will see that this yields post-classical sensitivity for multiparameter estimation even for an asymptotically large number of modes. This system is also considered for nondeterministic photon sources and a variety of measurement schemes.
Quantum number-path entanglement is a resource for supersensitive quantum metrology and in particular provides for sub-shot-noise or even Heisenberg-limited sensitivity. However, such number-path entanglement is thought to have been resource intensive to create in the first place, typically requiring either very strong nonlinearities or nondeterministic preparation schemes with feedforward, which are difficult to implement. Recently [K. R. Motes et al., Phys. Rev. Lett. 114, 170802 (2015)], it was shown that number-path entanglement from a BosonSampling inspired interferometer can be used to beat the shot-noise limit. In this paper we compare and contrast different interferometric schemes, discuss resource counting, calculate exact quantum Cram\'er-Rao bounds, and study details of experimental errors.
Abstract Estimating transmission or loss is at the heart of spectroscopy. To achieve the ultimate quantum resolution limit, one must use probe states with definite photon number and detectors capable of distinguishing the number of photons impinging thereon. In practice, one can outperform classical limits using two-mode squeezed light, which can be used to herald definite-photon-number probes, but the heralding is not guaranteed to produce the desired probes when there is loss in the heralding arm or its detector is imperfect. We show that this paradigm can be used to simultaneously measure distinct loss parameters in both modes of the squeezed light, with attainable quantum advantages. We demonstrate this protocol on Xanadu’s X8 chip, accessed via the cloud, building photon-number probability distributions from 106 shots and performing maximum likelihood estimation (MLE) on these distributions 103 independent times. Because pump light may be lost before the squeezing occurs, we also simultaneously estimate the actual input power, using the theory of nuisance parameters. MLE converges to estimate the transmission amplitudes in X8’s eight modes to be 0.392 02(6), 0.307 06(8), 0.369 37(6), 0.287 30(9), 0.382 06(6), 0.304 41(8), 0.372 29(6), and 0.286 21(8) and the squeezing parameters, which are proxies for effective input coherent-state amplitudes, their losses, and their nonlinear interaction times, to be 1.3000(2), 1.3238(3), 1.2666(2), and 1.3425(3); all of these uncertainties are within a factor of two of the quantum Cramér–Rao bound. This study provides crucial insight into the intersection of quantum multiparameter estimation theory, MLE convergence, and the characterization and performance of real quantum devices.
Abstract We propose a generalized form of entangled coherent states (ECS) and apply them in a multi-arm optical interferometer to estimate multiple phase shifts. We obtain the quantum Cramér–Rao bounds for both the linear and nonlinear parametrization protocols. Through analysis utilizing simultaneous estimation we find that this generalized form of ECS gives better precision than the generalized NOON states (2013 Phys. Rev. Lett. 111 070403 ). Moreover, when compared with the independent estimation both the linear and nonlinear protocols have the same advantage in the relation to the number of parameters.
The fundamental precision limit of an interferometer is crucial since it bounds the best possible sensitivity one could achieve using such a device. This thesis will focus on several different interferometers and try to give the ultimate precision bounds by carefully counting all the resources used in the interferometers. The thesis begins with the basics of the quantum state of light. The fundamentals of quantum metrology are also reviewed and discussed. More specifically, the terminology of classical and quantum Cram\'er-Rao bound and classical and quantum Fisher information are introduced. Chapter 3 discusses the conclusive precision bounds in two-mode interferometer such as Mach-Zehnder interferometer (MZI) and SU(1,1) interferometer. I revisit the quantum Fisher information approach of these two interferometers and show the discrepancy of phase sensitivity on the physically same setup. Then I establish fundamental precision estimation bounds for such device, due to the reason that many works of literature fail to accurately count the resources and knowledge of phase-to-be-estimated used in the interferometers. The analysis suggests that for a MZI, one can never do better than SNL in phase sensitivity, when an input to one of the two ports is the vacuum. If one does not allow the detector to use any external phase reference or power resource, then the precision is limited by the SNL. For a SU(1,1) interferometer, firstly, when one of the input states is restricted to be a vacuum state, I showed that by using either the phase-averaging method or the quantum Fisher information matrix method, different phase configurations of the SU(1,1) interferometer result in the same QFI. Secondly, I compared the results of the phase-averaging method and the quantum Fisher information matrix method, and then I argued that for an SU(1,1) interferometer, phase averaging or quantum Fisher information matrix method is generally required, and they are equivalent. Finally, I used the quantum Fisher information matrix method to calculate the precision limit for other common input states, such as two coherent state inputs or coherent state with squeezed vacuum inputs. In chapter 4, I will consider a passive multi-mode interferometer for multiparameter phase estimation. It was suggested that optical networks with relatively inexpensive overhead---single photon Fock states, passive optical elements, and single photon detection---can show significant improvements over classical strategies for single-parameter estimation, when the number of modes in the network is small. In this chapter, I analytically compute the quantum Cramer-Rao bound to show these
This paper explores multiparameter quantum metrology using Greenberger–Horne–Zeilinger (GHZ)-type photon-added coherent states (PACS) and investigates both independent and simultaneous parameter estimation with linear and nonlinear protocols, highlighting the significant potential of quantum resources to enhance precision in multiparameter scenarios. To provide a comprehensive analysis, we explicitly derive analytical expressions for the quantum Cramér–Rao bound (QCRB) for each protocol. Additionally, we compare the two estimation strategies, examining the behavior of their QCRBs and offering insights into the advantages and limitations of these quantum states in various contexts. Our results show that simultaneous estimation generally outperforms independent estimation, particularly in nonlinear protocols. Furthermore, we analyze how the QCRB varies with the coherent state amplitude [Formula: see text], the number of estimated parameters d, and the photon excitation order n across three protocols. The results indicate that increasing [Formula: see text] and decreasing d improves estimation precision. For low n, the variation in the QCRB is similar for both symmetric and antisymmetric cases; however, at higher n, the antisymmetric case exhibits slightly better precision. The dependence on d is comparable for both types of states. We also compare PACS-based GHZ states with NOON states and entangled coherent states, demonstrating the relative performance of each. Finally, we conclude with an analysis of homodyne detection in the context of a linear protocol, discussing its impact on estimation accuracy.