Quantum Fisher information and quadrature squeezing in Janus superpositions of squeezed vacua

Arash Azizi · 2026

Read Quantum Fisher information and quadrature squeezing in Janus superpositions of squeezed vacua by Arash Azizi (2026) 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-2603-15720.txt.

Introduction

Azizi studies Janus states, built by superposing two squeezed vacuum states. Such states are non-Gaussian, so plain squeezing rules may mislead. He computes their quantum Fisher information with exact analytic tools. At fixed mean photon number, a single squeezed vacuum stays best for basic squeezing. Inside a fixed two-state span, Janus states can beat both of their parts. They win on quadrature variance and on phase sensitivity at once. A fixed-squeezing benchmark shows still larger gains for quadratic sensing.

Catalog record

CREATORArash Azizi
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE101402
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.15720

Browse the full library · About the Tower of Babel

Full text

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

Quantum Fisher information and quadrature squeezing in Janus superpositions of squeezed vacua Arash Azizi Abstract Janus states, defined as coherent superpositions of two single-mode squeezed vacua, provide a simple but genuinely non-Gaussian setting for studying how interference reshapes quantum Fisher information (QFI) beyond the Gaussian squeezed-vacuum picture. Using an exact analytic treatment, we determine the QFI of Janus states and identify the benchmarks under which they can or cannot offer a metrological advantage over the single squeezed vacuum. We find that, under a fair comparison at fixed mean photon number, the single squeezed vacuum remains optimal for principal second-moment squeezing, so no genuine Janus advantage exists at that level. By contrast, within a fixed two-state span, a Janus superposition can simultaneously outperform its constituents in a laboratory quadrature variance and in number-generated phase QFI. We also introduce an operational benchmark based on fixed measured squeezing and show that, at the same observed squeezing level, Janus interference can substantially enhance the QFI for quadratic-generator sensing beyond the pure-Gaussian squeezed-vacuum reference. These results show that the metrological performance of Janus states is controlled not only by quadrature squeezing, but also by higher-order fluctuations and by the benchmark used for comparison. I Introduction Precision sensing lies at the heart of modern quantum science. In a typical metrological task, one seeks to estimate a small physical parameter—such as a phase, frequency, coupling strength, displacement, or loss coefficient—from measurements on a probe whose quantum state depends on that parameter. Quantum metrology provides the framework for this problem by identifying the ultimate precision allowed by quantum mechanics and by clarifying when nonclassical probes can outperform classical ones under a specified resource constraint 8 ; 15 ; 16 ; 24 ; 25 ; 31 ; 12 ; 7 . In the pure-state unitary setting, the central figure of merit is the quantum Fisher information (QFI), which is equal to four times the variance of the generator that encodes the parameter. The QFI therefore serves both as a rigorous precision benchmark and as a practical language for comparing probe families, sensing generators, and resource costs. This perspective is especially natural in quantum optics, where the probe is often a bosonic mode and the relevant fluctuations can be shaped through squeezing, interference, and non-Gaussian superposition 19 ; 29 ; 18 ; 14 . The canonical Gaussian resource is the single-mode squeezed vacuum, whose reduced quadrature noise has become a standard tool in precision measurement 11 ; 34 ; 36 ; 17 ; 26 . Experimentally, squeezed light has advanced from early optical demonstrations to high-purity modern platforms and large-scale interferometric implementations 30 ; 35 ; 9 ; 32 ; 20 ; 33 ; 1 ; 23 ; 2 ; 28 . At the same time, these developments have made an important point increasingly clear: reduced quadrature noise, nonclassical photon correlations, and enhanced QFI are closely related, but they are not identical notions of metrological advantage. Superpositions of squeezed states have long been part of the quantum-optical landscape 27 ; 10 ; 22 ; 21 ; 13 . The Janus program revisited this class from a more specific perspective, namely the role of exact optical coherence in non-Gaussian interference. The original Janus work derived an analytic expression for g^{(2)} for a coherent superposition of two squeezed vacua and showed that interference can strongly suppress two-photon events while still obeying the universal lower bound g^{(2)}\geq 1/2 , with a practical minimum reached at moderate squeezing 6 . A later study extended this analysis to arbitrary-order coherence by deriving exact expressions for g^{(k)} , thereby showing that Janus interference can act as a tunable switch for multiphoton correlations through a squeezing-polynomial framework 3 . The displaced Janus generalization then broadened the construction to superpositions of squeezed coherent states with a common displacement and developed a wider analytical toolbox that included factorial moments, Wigner functions, non-Gaussianity diagnostics, and QFI for linear and quadratic generators 4 . More recently, the same program was extended to the two-mode setting through the Two-Mode Janus State (TMJS), defined as a coherent superposition of two distinct two-mode squeezed states and developed as a direct non-Gaussian generalization of the thermofield double, with exact phase-steerable higher-order coherence and a corresponding two-mode squeezing-polynomial structure 5 . Those results established the coherence-theoretic and non-Gaussian foundations of the Janus family, but they did not settle the metrological question studied here. Once the discussion moves from g^{(2)}(0) and g^{(k)}(0) to QFI, different issues become central. One must distinguish between fixed-axis and principal-axis squeezing, between second-moment suppression and higher-order fluctuation enhancement, and between benchmarks that are mathematically convenient and those that correspond to what is actually measured in the laboratory. The central question is therefore not simply whether Janus interference modifies noise or correlations, but under which physically meaningful comparison it yields a genuine metrological advantage over the single squeezed vacuum. The purpose of the present work is to answer that question for Janus superpositions of two single-mode squeezed vacua by focusing directly on quadrature structure and QFI. We first develop the second-moment description in a form suited to metrology. In particular, we derive explicit closed expressions for the laboratory-axis quadrature variances, for the principal squeezing selected by homodyne optimization, and for the smallest fixed-axis variance attainable when the state is optimized only within a prescribed two-state squeezed-vacuum span. This analysis already reveals a key distinction: the minimum quadrature noise of a given state and the minimum fixed-axis noise achievable within a chosen Janus span are different optimization problems, and they lead to different notions of squeezing advantage. We then connect this quadrature analysis to explicit QFI benchmarks. Here the situation is subtler than the earlier Janus correlation studies might suggest. A state with lower laboratory-axis noise does not necessarily possess a larger QFI, and a state with strong photon-correlation signatures does not necessarily outperform the Gaussian squeezed-vacuum reference once the resource comparison is made fairly. For this reason, we analyze Janus probes under several distinct benchmarks. For number-generated phase estimation, we compare Janus states with the single squeezed vacuum at fixed mean photon number and determine when Janus interference can enhance the phase QFI. For quadratic-generator sensing, we introduce a benchmark based on fixed measured squeezing, which is especially natural experimentally because squeezing is often reported directly in dB through homodyne measurements. This allows one to ask whether two probes with the same observed squeezing can nevertheless possess substantially different metrological power. The main conclusion is that the answer depends decisively on the benchmark. Under a fair fixed-mean-photon-number comparison, the single squeezed vacuum remains extremal for principal second-moment squeezing, so no genuine Janus advantage survives at that level. By contrast, within a prescribed two-state squeezed-vacuum span, Janus interference can reduce a fixed laboratory-axis variance below that of either constituent. The same benchmark sensitivity appears at the level of QFI. For number-generated phase estimation, a Janus state can outperform the single squeezed vacuum at fixed mean photon number only through the higher-order fluctuation structure encoded in its factorial moments. For quadratic-generator sensing, however, substantial enhancement relative to the pure-Gaussian reference can occur when the comparison is made at fixed measured squeezing. The point of the present work is therefore not merely to ask whether Janus

Read the complete text