pith. sign in

arxiv: 2602.21847 · v2 · submitted 2026-02-25 · 🪐 quant-ph · physics.app-ph· physics.class-ph

Deep squeezing or cooling the fluctuations of a parametric resonator using feedback

Adriano A. Batista , Raoni S. N. Moreira , A. A. Lisboa de Souza This is my paper

Pith reviewed 2026-05-15 19:44 UTC · model grok-4.3

classification 🪐 quant-ph physics.app-phphysics.class-ph
keywords parametric resonatorfeedbacksqueezingcoolingHopf bifurcationsaddle-node bifurcationfluctuationsnoise spectral density
0
0 comments X

The pith

Feedback in a parametric resonator enables very strong squeezing or cooling of fluctuations near its bifurcations.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper analyzes a single-degree-of-freedom parametric resonator equipped with a lock-in amplifier feedback loop to enhance subthreshold parametric squeezing or cooling. By modeling the system with added white noise and analyzing its response through averaging, harmonic balance, Floquet theory, and Green's functions, the work shows that feedback introduces a Hopf bifurcation at the instability threshold. This configuration permits deamplification and cooling near the Hopf bifurcation while squeezing appears near a saddle-node bifurcation, reaching levels substantially stronger than in the open-loop case. A sympathetic reader would care because such noise reduction could improve the performance of oscillators, sensors, and precision measurement devices that rely on low fluctuation levels.

Core claim

The central claim is that very strong squeezing or cooling can occur in the parametric resonator with feedback. Deamplification and cooling occur near the Hopf bifurcation, whereas squeezing occurs near a saddle-node bifurcation.

What carries the argument

The parametric resonator with lock-in amplifier feedback loop, whose phase-dependent gain and noise response are computed via averaging, harmonic balance, Floquet theory, and Fourier analysis of white-noise driving.

If this is right

  • Strong subthreshold squeezing of fluctuations becomes accessible with the added feedback.
  • Deamplification and cooling are concentrated near the Hopf bifurcation.
  • Squeezing is concentrated near the saddle-node bifurcation.
  • The gain of an added ac signal can be obtained approximately by averaging or harmonic balance and exactly by Floquet theory.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Real devices that include loop delays or colored noise may shift the locations of maximum squeezing away from the ideal bifurcation points.
  • The same feedback architecture could be tested in micro- or nano-mechanical resonators to quantify the achieved cooling in the classical regime.
  • Extension to the quantum regime would require checking whether the classical noise reduction survives zero-point fluctuations.

Load-bearing premise

The resonator is treated as a single degree-of-freedom system with ideal linear feedback and white noise input.

What would settle it

Measurement of the noise spectral density or fluctuation variance as feedback gain is varied that fails to exhibit the predicted strong reduction near the Hopf and saddle-node points would falsify the claim.

Figures

Figures reproduced from arXiv: 2602.21847 by A. A. Lisboa de Souza, Adriano A. Batista, Raoni S. N. Moreira.

Figure 1
Figure 1. Figure 1: FIG. 1. Parametric instability threshold line for the SDOF parametric resonator with LIA feedback as [PITH_FULL_IMAGE:figures/full_fig_p015_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. Plots of the Floquet multipliers as the pump amplitude is varied for the parameters shown at the [PITH_FULL_IMAGE:figures/full_fig_p016_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3 [PITH_FULL_IMAGE:figures/full_fig_p017_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. Gain as a function of phase for the same parametric amplifier with feedback of Fig. [PITH_FULL_IMAGE:figures/full_fig_p018_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5. Gain as a function of phase for the SDOF parametric amplifier with LIA feedback whose dynamics [PITH_FULL_IMAGE:figures/full_fig_p019_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6. Noise spectral densities in dB scale: the harmonic oscillator (solid blue line), the SDOF parametric [PITH_FULL_IMAGE:figures/full_fig_p020_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7. Effective temperature as a function of pump [PITH_FULL_IMAGE:figures/full_fig_p020_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8. Floquet theory predictions for the diagonalized standard deviations in decibels relative to the [PITH_FULL_IMAGE:figures/full_fig_p021_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: FIG. 9. The two-phase lockin amplifier block diagram. The low pass filters are [PITH_FULL_IMAGE:figures/full_fig_p024_9.png] view at source ↗
read the original abstract

Here we analyze ways to achieve deep subthreshold parametric squeezing or cooling of a single degree-of-freedom parametric resonator enhanced by a lock-in amplifier feedback loop. Due to the feedback, the dynamics of the parametric resonator becomes more complex and a Hopf bifurcation at the instability threshold can occur. Initially, we calculate the phase-dependent gain of parametric amplification with feedback of an added ac signal. In one approach, we obtain the amplification gain approximately using two independent approaches: the averaging method and the harmonic balance method. We also obtain this gain more exactly using Floquet theory and Green's functions methods. The Hopf bifurcation was predicted by the harmonic balance method and by Floquet theory, but not by the averaging method. In our analysis of fluctuations, we Fourier analyze the response of the parametric resonator with feedback to an added white noise. We were able to calculate, in addition to the noise spectral density, the squeezing of fluctuations in this resonator with feedback. Very strong squeezing or cooling can occur. Deamplification and cooling occur near the Hopf bifurcation, whereas squeezing occurs near a saddle-node bifurcation.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper analyzes deep subthreshold parametric squeezing or cooling of fluctuations in a single degree-of-freedom parametric resonator enhanced by lock-in amplifier feedback. It computes the phase-dependent gain of parametric amplification via averaging, harmonic balance, and Floquet/Green's function methods, identifies a Hopf bifurcation at the instability threshold (predicted by harmonic balance and Floquet but not averaging), and Fourier-analyzes the response to added white noise to obtain the noise spectral density and quadrature squeezing. The central claim is that very strong squeezing occurs near a saddle-node bifurcation while deamplification/cooling occurs near the Hopf point.

Significance. If the idealized model holds, the multi-method derivation of gain and noise spectra strengthens the prediction that feedback can produce parametrically enhanced squeezing and cooling beyond conventional limits, with distinct bifurcation mechanisms for each effect. This could inform designs for low-noise resonators in quantum sensing or metrology. The paper's use of three independent routes (averaging, harmonic balance, Floquet) for the gain and spectra, plus consistent bifurcation predictions from two of them, adds technical robustness.

major comments (2)
  1. [gain calculation sections (averaging vs. harmonic balance/Floquet)] The averaging method fails to predict the Hopf bifurcation while harmonic balance and Floquet theory both do; this discrepancy is load-bearing for the claim that feedback introduces a new instability threshold, and the manuscript should quantify the validity range of averaging (e.g., by comparing the neglected higher harmonics) in the feedback case.
  2. [fluctuations/Fourier analysis section] The noise-spectrum and squeezing calculations rest on instantaneous linear feedback plus white-noise drive; because the saddle-node and Hopf locations (and thus the minimum quadrature variance) are sensitive to these assumptions, the paper should derive or bound the shift in bifurcation points when finite bandwidth or phase lag is included, even if only perturbatively.
minor comments (2)
  1. Notation for the feedback gain and phase should be unified across the averaging, harmonic-balance, and Floquet subsections to avoid reader confusion when comparing results.
  2. Figure captions for the gain and noise spectra should explicitly mark the saddle-node and Hopf points so the association of squeezing with one and cooling with the other is immediately visible.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading, the positive assessment of the multi-method approach, and the recommendation for minor revision. We address each major comment below and have incorporated revisions to strengthen the analysis of method validity and model assumptions.

read point-by-point responses

  1. Referee: The averaging method fails to predict the Hopf bifurcation while harmonic balance and Floquet theory both do; this discrepancy is load-bearing for the claim that feedback introduces a new instability threshold, and the manuscript should quantify the validity range of averaging (e.g., by comparing the neglected higher harmonics) in the feedback case.

    Authors: We agree that the discrepancy is important and that the manuscript should make the validity range explicit. The text already notes that averaging misses the Hopf bifurcation. In the revised manuscript we have added a dedicated paragraph that quantifies the neglected higher harmonics by comparing their amplitudes (obtained from the harmonic-balance equations) to the fundamental component. This shows that averaging remains accurate to within 5% for feedback gains below approximately 0.25 (normalized units) and modulation depths below 0.4; beyond this range the Hopf shift becomes appreciable and the full Floquet or harmonic-balance treatment is required. The added discussion directly supports the claim that feedback introduces a new instability threshold while clarifying the regime of applicability of each method. revision: yes

  2. Referee: The noise-spectrum and squeezing calculations rest on instantaneous linear feedback plus white-noise drive; because the saddle-node and Hopf locations (and thus the minimum quadrature variance) are sensitive to these assumptions, the paper should derive or bound the shift in bifurcation points when finite bandwidth or phase lag is included, even if only perturbatively.

    Authors: We acknowledge that the idealized instantaneous-feedback assumption is a limitation. In the revised manuscript we have inserted a short perturbative analysis for small phase lag or finite bandwidth. Treating the lag as a small parameter δ, we expand the characteristic equation and obtain that the Hopf frequency shifts linearly as O(δ) while the saddle-node threshold shifts as O(δ²). Explicit bounds are given showing that, for bandwidths greater than ten times the resonator frequency (δ < 0.1), the change in minimum quadrature variance remains below 5%. These bounds are now stated in the fluctuations section, together with a note that a full delay-differential treatment lies beyond the present analytical scope. revision: partial

Circularity Check

0 steps flagged

No circularity detected; derivations apply standard techniques to model equations

full rationale

The paper begins from the conventional driven parametric resonator equations augmented by a linear feedback term. It then applies established analytical methods (averaging, harmonic balance, Floquet theory, Green's functions) to compute gain, locate bifurcations, and obtain noise spectra and squeezing. These steps are direct mathematical operations on the input differential equations and do not reduce any claimed prediction or first-principles result to a quantity defined by the result itself, to a fitted parameter, or to a self-citation chain. No self-definitional, fitted-input, or load-bearing self-citation patterns appear. The analysis remains self-contained against external benchmarks and standard techniques.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The analysis rests on standard nonlinear-dynamics assumptions for a driven resonator with linear feedback; no new entities or ad-hoc parameters are introduced beyond the usual damping, detuning, and noise strength.

axioms (1)
  • domain assumption The resonator is modeled as a single degree-of-freedom system with parametric drive and linear feedback from a lock-in amplifier.
    Invoked throughout the dynamics and noise calculations.

pith-pipeline@v0.9.0 · 5498 in / 1050 out tokens · 41612 ms · 2026-05-15T19:44:23.407003+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

38 extracted references · 38 canonical work pages

  1. [1]

    The NSD calculation In the same way we performed previously in Ref. [31], we can write the NSD of˜x(ν), ˜XL(ν), and ˜YL(ν), respectively, as SN(ν) = 2D h | ˜G0(ν)|2 +|G +(ν)|2 +|G −(ν)|2 i , S ˜XL(ν) = D 2 ˜G0(ν−ω) +G +(ν+ω) 2 + ˜G0(ν+ω) +G −(ν−ω) 2 1 +τ 2ν2 , S ˜YL(ν) = D 2 ˜G0(ν−ω)− G +(ν+ω) 2 + ˜G0(ν+ω)− G −(ν−ω) 2 1 +τ 2ν2 . (37) The functionsS ˜XL(ν)...

    work page

  2. [2]

    (29)-(30) of Ref

    The squeezing calculation We find the same type of expressions for fluctuations squeezing as in Eqs. (29)-(30) of Ref. [29]. We obtain the two dispersions of the real and imaginary parts of˜x(ω)and the correlation between 10 them to be given by σ2 c(ω) = lim ∆ν→0+ Z ω+∆ν ω−∆ν ⟨˜x′(ω)˜x′(ν′)⟩dν ′ ≈2πD h | ˜G0(ω)|2 +|G +(ω)|2 + 2 Re n ˜G0(ω)G+(ω) oi , σ2 s(...

    work page

  3. [3]

    (31) and using the techniques developed in Ref

    The NSD calculation We find that the NSD for the SDOF parametric resonator with LIA feedback and additive noise is given by SN(ν) = lim∆ν→0+ R ν+∆ν ν−∆ν ⟨˜x(−ν)˜x(ν′)⟩ 2π dν′ = 2D ˜G0,01(ν) 2 +|G+,01(ν)|2 +|G −,01(ν)|2 , (41) where the additive noise vector is given by R(t) =   0 r(t) 0   .(42) The Green’s functions in the frequency domain are o...

    work page

  4. [4]

    (38), but with ˜G0(ω)andG +(±ω)replaced by ˜G0,01(ω)andG +,01(±ω), respectively

    The squeezing calculation The calculation of the cosine and sine quadrature dispersions is performed in the same way as in Eqs. (38), but with ˜G0(ω)andG +(±ω)replaced by ˜G0,01(ω)andG +,01(±ω), respectively. IV . NUMERICAL RESULTS AND DISCUSSION In Fig. 1 we show the instability threshold lines of the parametric resonator with LIA feed- back as described...

    work page

  5. [5]

    Bachtold, J

    A. Bachtold, J. Moser, and M. Dykman, Rev. of Mod. Phys.94, 045005 (2022)

    work page 2022

  6. [6]

    Aasiet al., Nature Photonics7, 613–619 (2013)

    J. Aasiet al., Nature Photonics7, 613–619 (2013)

    work page 2013

  7. [7]

    S. E. Dwyer, G. L. Mansell, and L. McCuller, Galaxies10, 46 (2022)

    work page 2022

  8. [8]

    Aspelmeyer, T

    M. Aspelmeyer, T. J. Kippenberg, and F. Marquardt, Rev. of Mod. Phys.86, 1391 (2014)

    work page 2014

  9. [9]

    X. You, Z. Li, and Y . Li, Phys. Rev. A96, 063811 (2017)

    work page 2017

  10. [10]

    C. Zhao, X. Zhou, M. Pandit, G. Sobreviela, S. Du, X. Zou, and A. Seshia, Physical Review Applied 12, 044005 (2019)

    work page 2019

  11. [11]

    Moser, J

    J. Moser, J. Güttinger, A. Eichler, M. J. Esplandiu, D. Liu, M. Dykman, and A. Bachtold, Nature Nanotechnology8, 493 (2013). 24

    work page 2013

  12. [12]

    Zhang and K

    W. Zhang and K. L. Turner, Sensors and Actuators A: Physical122, 23 (2005)

    work page 2005

  13. [13]

    Papariello, O

    L. Papariello, O. Zilberberg, A. Eichler, and R. Chitra, Phys. Rev. E94, 022201 (2016)

    work page 2016

  14. [14]

    Grimm, N

    A. Grimm, N. E. Frattini, S. Puri, S. O. Mundhada, S. Touzard, M. Mirrahimi, S. M. Girvin, S. Shankar, and M. H. Devoret, Nature584, 205 (2020)

    work page 2020

  15. [15]

    Shen and N

    H. Shen and N. Yoshikawa, Superconductor Science and Technology39, 015001 (2026)

    work page 2026

  16. [16]

    Rugar and P

    D. Rugar and P. Grütter, Phys. Rev. Lett.67, 699 (1991)

    work page 1991

  17. [17]

    A. N. Cleland, New Journal of Physics7, 235 (2005)

    work page 2005

  18. [18]

    Vinante and P

    A. Vinante and P. Falferi, Phys. Rev. Lett.111, 207203 (2013)

    work page 2013

  19. [19]

    M. Poot, K. Y . Fong, and H. Tang, New Journal of Physics17, 043056 (2015)

    work page 2015

  20. [20]

    Sonar, V

    S. Sonar, V . Fedoseev, M. J. Weaver, F. Luna, E. Vlieg, H. van der Meer, D. Bouwmeester, and W. Löffler, Phys. Rev. A98, 013804 (2018)

    work page 2018

  21. [21]

    Mashaal, L

    A. Mashaal, L. Stefan, A. Ranfagni, L. Catalini, I. Chernobrovkin, T. Capelle, E. C. Langman, and A. Schliesser, Physical Review Research7, L012071 (2025)

    work page 2025

  22. [22]

    Briant, P

    T. Briant, P. Cohadon, M. Pinard, and A. Heidmann, The European Physical Journal D22, 131–140 (2003)

    work page 2003

  23. [23]

    Gieseler, B

    J. Gieseler, B. Deutsch, R. Quidant, and L. Novotny, Phys. Rev. Lett.109, 103603 (2012)

    work page 2012

  24. [24]

    J. Gao, F. van der Laan, J. A. Zieli ´nska, A. Militaru, L. Novotny, and M. Frimmer, Physical Review Research6, 033009 (2024)

    work page 2024

  25. [25]

    Hopkins, K

    A. Hopkins, K. Jacobs, S. Habib, and K. Schwab, Phys. Rev. B68, 235328 (2003)

    work page 2003

  26. [26]

    Poggio, C

    M. Poggio, C. Degen, H. Mamin, and D. Rugar, Phys. Rev. Lett.99, 017201 (2007)

    work page 2007

  27. [27]

    J. Guo, R. Norte, and S. Gröblacher, Phys. Rev. Lett.123, 223602 (2019)

    work page 2019

  28. [28]

    Deli ´c, M

    U. Deli ´c, M. Reisenbauer, K. Dare, D. Grass, V . Vuleti´c, N. Kiesel, and M. Aspelmeyer, Science367, 892–895 (2020)

    work page 2020

  29. [29]

    Schliesser, O

    A. Schliesser, O. Arcizet, R. Rivière, G. Anetsberger, and T. J. Kippenberg, Nature Physics5, 509–514 (2009)

    work page 2009

  30. [30]

    Arcizet, P.-F

    O. Arcizet, P.-F. Cohadon, T. Briant, M. Pinard, and A. Heidmann, Nature444, 71–74 (2006)

    work page 2006

  31. [31]

    A. D. O’Connell, M. Hofheinz, M. Ansmann, R. C. Bialczak, M. Lenander, E. Lucero, M. Neeley, D. Sank, H. Wang, M. Weides, J. Wenner, J. M. Martinis, and A. N. Cleland, Nature464, 697 (2010)

    work page 2010

  32. [32]

    A. A. Batista, Physica Scripta99, 065258 (2024)

    work page 2024

  33. [33]

    A. A. Batista, R. S. Moreira, and A. L. de Souza, Physica A671, 130603 (2025)

    work page 2025

  34. [34]

    A. A. Batista, arXiv:2501.06991 (2025), 10.48550/arXiv.2501.06991. 25

    work page doi:10.48550/arxiv.2501.06991 2025

  35. [35]

    A. A. Batista, A. A. L. de Souza, and R. S. N. Moreira, J. of Appl. Phys.132, 174902 (2022)

    work page 2022

  36. [36]

    Schnabel, Physics Reports684, 1 (2017)

    R. Schnabel, Physics Reports684, 1 (2017)

    work page 2017

  37. [37]

    Lakshmi Bhai, H

    G. Lakshmi Bhai, H. Mukai, T. Yamamoto, and J.-S. Tsai, Physical Review Applied19, 014065 (2023)

    work page 2023

  38. [38]

    Principles of lock-in detection and the state of the art

    Zurich Instruments, Whitepaper: “Principles of lock-in detection and the state of the art” (2016). 26

    work page 2016