Scalable Network of Mach-Zehnder Interferometers with a Single Entangled Resource

Zhihui Yan 1,2, Yanni Feng 1, Luca Pezze 3, Zhaoqing Zeng 1, Jingxu Ma 1, Xiaoyu Zhou 1, Augusto Smerzi 3, Xiaojun Jia 1,2, Kunchi Peng 1,2 · 2025

Read Scalable Network of Mach-Zehnder Interferometers with a Single Entangled Resource by Zhihui Yan 1,2, Yanni Feng 1, Luca Pezze 3, Zhaoqing Zeng 1, Jingxu Ma 1, Xiaoyu Zhou 1, Augusto Smerzi 3, Xiaojun Jia 1,2, Kunchi Peng 1,2 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-08230.txt.

Introduction

One shared squeezed-light source entangles a whole network of interferometers in a 2025 quantum sensing paper by Zhihui Yan and 8 co-authors. Distributed quantum sensing promises precision beyond classical limits for navigation, imaging, and environmental monitoring, but distributing entanglement optimally across many nodes is complex and fragile.

The demonstrated network links Mach-Zehnder interferometers through a single entangled resource, and it can be reconfigured as needs change. Simplicity is the point: one resource serving many sensors scales better than bespoke entanglement plumbing per node. Sensing several quantities at once, across separated sites, with quantum-enhanced precision is a capability laboratories have chased for years, and a reconfigurable single-source design moves it closer to practice.

Catalog record

CREATORZhihui Yan 1,2, Yanni Feng 1, Luca Pezze 3, Zhaoqing Zeng 1, Jingxu Ma 1, Xiaoyu Zhou 1, Augusto Smerzi 3, Xiaojun Jia 1,2, Kunchi Peng 1,2
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE58 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-08230

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Scalable Network of Mach-Zehnder Interferometers with a Single Entangled Resource Zhihui Yan 1,2, Yanni Feng 1, Luca Pezze 3, Zhaoqing Zeng 1, Jingxu Ma 1, Xiaoyu Zhou 1, Augusto Smerzi 3, Xiaojun Jia 1,2, Kunchi Peng 1,2 Abstract Distributed quantum sensing exploits entanglement to enhance the estimation of multiple parameters across a network of spatially-separated sensors, achieving sensitivities beyond the classical limit. Potential applications cover a plethora of technologies, from precision navigation to biomedical imaging and environmental monitoring. However, practical implementations are challenged by the complex optimal distribution of entanglement throughout the sensing nodes, which affects scalability and robustness. Here we demonstrate a reconfigurable network of Mach–Zehnder interferometers entangled via a single shared squeezed-vacuum resource. We achieve joint noise suppression of 4.36 ± 0.35 4.36\pm 0.35 dB below the standard quantum limit at the phase-uncertainty level of 10 − 9 10^{-9} . Furthermore, after full optimization in the low-intensity regime, we demonstrate a crossover from the standard quantum limit to the Heisenberg limit. The network estimates arbitrary linear combinations of phases, saturates the quantum Cramér-Rao bound in the ideal case, remains robust under realistic photon losses, and scales favorably with the number of sensors. Quantum interferometers are the most sensitive measurement devices in modern science and technology, with applications ranging from gravitational-wave detection AbbottPRL2016 to nanoscale imaging CasacioNATURE2021 . A foundational example is the Mach-Zehnder interferometer (MZI), which serves as a key platform in both fundamental studies and practical implementations PirandolaNATPHOT2018 . The MZI offers enhanced phase estimation by exploiting high-intensity optical fields and common-mode noise rejection, with a precision fundamentally limited by quantum fluctuations SchnabelNATCOMM2010 , LawrieACSP2019 , PolinoAVSQS2020 . While conventional MZIs operate at the standard quantum limit (SQL) PezzePRL2007 , sub-SQL sensitivities have been demonstrated by injecting nonclassical states – such as squeezed or entangled light CavesPRD1981 , XiaoPRL1987 , HollandPRL1993 , PezzePRL2008 , JooPRL2011 , LangPRL2013 – into the interferometer. These advancements have been realized in proof-of-principle experiments NagataSCIENCE2007 , AfekSCIENCE2010 , SlussarenkoNATPHOT2017 , DaryanooshNATCOMM2018 , ZuoPRL2020 , NielsenPRL2023 as well as in large-scale observatories such as Advanced LIGO and Virgo GodaNATPHYS2008 , AbadieNATPHYS2011 , AasiNATPHOT2013 , TsePRL2019 , AcernesePRL2019 , JiaSCIENCE2024 . Most quantum-enhanced interferometric implementations remain focused on estimating a single phase shift. However, a wide range of emerging applications – including biological imaging Juan , Hoffmann , DefienneNATPHOT2024 , inertial navigation GravePRAPP2020 and environmental monitoring G2022 – requires the simultaneous estimation of multiple phases across a network of distributed sensing nodes. To address this need, entanglement-enhanced distributed quantum sensing (DQS) is rapidly gaining prominence as a key quantum technology PezzeARXIV2025 . Figure 1: Distributed quantum sensing with a scalable entangled network of Mach-Zehnder interferometers. In this schematic architecture, the central node (green cylinder) is responsible for generating and distributing mode entanglement across spatially separated MZIs (orange). The node contains a squeezer (red ellipse), a displacer (blue disc), and a quantum circuit. Each local MZI receives two input modes: a coherent mode (blue arrow) and an entangled mode (red arrow): mode entanglement generated by the multimode splitting of a single squeezed-vacuum state is represented by the red halo. Within each MZI, the two inputs are combined at a beam splitter, acquire a relative phase shift (the zig-zag lines indicating multipass phase amplification), and are then recombined at a second beam splitter before detection. Measurement outcomes are collected from a single output port of each MZI. The wavy background schematically represents an inhomogeneous field. This field couples to the interferometers by inducing local relative phase shifts between the two sensing arms. Figure 2: Experimental realization of an entangled network of quantum interferometers. a Schematic of the experimental apparatus. A single squeezed-vacuum state is generated by an optical parametric amplifier (OPA) and evenly distributed via a quantum circuit (QC) composed of polarization beam splitters. The remaining d-1 inputs to the QC are vacuum states (with d=6 in this implementation). Linear splitting of the squeezed input generates multipartite entanglement across the d output modes, labeled b_{1},...,b_{d} . Each output mode b_{j} is injected into the j th MZI (thin arrow), while the second input port a_{j} receives a coherent state (thick arrow). Each MZI comprises a balanced beam splitter, a phase interrogation region where the beam undergoes K passes through the sample, and a final balanced beam splitter. The phase shift accrued in a single-pass through the j th interferometer is denoted by \theta_{j} . The output mode \tilde{b}_{j} (thin line) is analyzed by a local balanced homodyne detection (BHD), using an external local oscillator (thick arrow). Multiphase estimation is achieved by jointly processing all BHD measurement outcomes. (Inset) Detail of the multipass interaction. PPKTP: periodically poled KTiOPO 4 ; HR: high-reflectivity mirror; HWP: half-wave plate; PBS: polarization beam splitter; BS: beam splitter; AOM: acousto-optic modulator; PZT: piezoelectric transducer; PD: photodetector; ATT: optical attenuator; MC: mode cleaner; SHG: second harmonic generator. Panels b and c show the joint relative noise powers of the phase average (\theta_{1}+...+\theta_{d})/d . The black solid line corresponds to the conventional network of MZI (namely, with no squeezing and single-pass interaction, K=1 ). It is compared with the noise powers obtained with the entangled network of MZIs (colored line) for single-pass (panel b ) and five-pass (panel c ) interactions. In spectrum analyzer, analyzing frequency is 4 MHz; resolution bandwidth is 100 kHz; video bandwidth is 1 kHz. In this manuscript, we realize experimentally a novel DQS optical architecture consisting of a scalable and versatile network of MZIs. The multiparameter sensor is schematically shown in Fig. 1 and builds on recent theoretical proposals 11 , PezzeARXIV . The scheme estimates arbitrary linear combinations of the relative phase shift in each MZI through optimal local measurements. The splitting of a single Gaussian squeezed-vacuum state generates entanglement among multiple modes. Each entangled mode enters one input port of a MZI, while a coherent state enters its second port. The hybrid quantum-classical nature of the probe state is a distinguishing trait of our scheme. Specifically, we exploit intense coherent states to reach phase uncertainties on the order of 10^{-9} with the total input power of 9.6 mW , achieving more than 4 dB below the SQL thanks to the distributed squeezed vacuum. We report absolute quantum-enhanced phase sensitivities that are orders of magnitude beyond those achieved in the current literature on multiparameter phase sensing. We further extend the approach to the opposite regime of low photon-intensity and optimize the distribution of photons between the split squeezed vacuum and the coherent states. In particular, we address a crossover regime between the SQL – which is proved optimal in the limit of vanishing intensities – and the Heisenberg limit, and demonstrate a sub-SQL sensitivity with respect to the total average photon number in the sensor. Unlike prior multiphase estimation schemes LiuNATPHOT2021 , CiminiAP2023 , HongNATCOMM2021 , KimNATCOMM2024 , ZhaoPRX2021 , our approach makes use of Gaussian states which prove inherently robust. Furthermore, our sensor network exploits the notable practical advantages of MZIs – offering noise-resilient differential measurements, lack of external phase reference and reduced sensitivity to local oscillator and alignment noise – for multi-phase estimation beyond homodyne GuoNATPHYS2020 and quadrature XiaPRL2020 sensing schemes. We directly demonstrate the protocol’s resilience to photon losses and show that a single squeezed-vacuum

Read the complete text