REVIEW 4 major objections 5 minor 36 references
A plug-and-play solution for characterizing two-way optical frequency transfer over free-space
T0 review · 4 major / 5 minor · reviewed 2026-08-09 · deepseek-v4-flash
Pith's one-line read A two-way free-space optical link hits 2e-19 fractional frequency instability at 10 seconds of averaging.
desk verdict Solid engineering demonstration of a two-way free-space link characterizer, but the headline 2e-19 is a best-run technical floor on a fully folded link, not a point-to-point atmospheric non-reciprocity measurement. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central object is the two-way free-space link characterization setup: a Michelson-type interferometer with acousto-optic modulators (AOMs) and fiber-optic retroreflectors, connected to a free-space path under test. The AOM shifts $\nu_{AOM1}$, $\nu_{AOM2}$ mimic separate lasers in a real point-to-point comparison, and the three beat notes A, B, C on each photodetector allow pairwise combinations. The load-bearing identity is Eq. (2), $(PD1_A-PD1_B)-(PD2_A-PD2_B)=(\Delta\nu^{21}_{fi}-\Delta\nu^{12}_{fi})-(\Delta\nu^{21}_{fs}-\Delta\nu^{12}_{fs})$, which cancels all interferometric noise terms and isolates the residual non-reciprocity of the free-space link. The fully folded link, formed by a fiber-optic retroreflector with an AOM at the far site, makes forward and backward beams traverse exactly the same air path so delay noise cancels; this turns the measurement into a characterization of the technical noise floor.
What would settle it
Replace the fiber-optic retroreflector at the far site with two separate terminals at both ends of a 3.4 km point-to-point link and measure the same beat-note combination. If the residual non-reciprocity stays at the $10^{-19}$ level at 10 s under comparable turbulence, the folded-link result represents a true point-to-point capability; if it rises by orders of magnitude toward the calculated delay limit, the reported number is specific to the folded geometry.
Extended reading notes
Core claim
The central claim is that a two-way continuous-wave optical frequency transfer scheme, free of interferometric noise and laser phase noise through post-processing, can measure the residual non-reciprocity of a free-space connection at a level compatible with optical clock comparisons. Concretely, combining the A and B beat notes from two balanced photodetectors according to Eq. (2) cancels all common-mode interferometric noise and leaves the difference of the free-space path noises; with the fully folded link the delay-induced part cancels exactly. The measured residual instability is $2.0 \times 10^{-19}$ at 10 s, reaching $8.5 \times 10^{-21}$ at 100 s, and the fractional frequency offsets are compatible with zero. The paper identifies shot noise as the short-term limit and AM-PM conversion through the detection electronics as the long-term limit.
Load-bearing premise
The result depends on the fully folded link making the forward and return beams traverse exactly the same air path, so that delay-induced non-reciprocity cancels and the measured $2.0 \times 10^{-19}$ is a technical noise floor rather than the atmospheric non-reciprocity of a separated point-to-point link.
Editorial extensions
If this is right
- The system can serve as a testbed for free-space optical clock links: at the reported residual non-reciprocity, 10-second cycle-slip-free intervals are already sufficient for accurate clock comparisons.
- The identified limiting factors give a concrete improvement path: reducing AM-PM conversion in the detection electronics and increasing received optical power to lower shot noise would push the technical floor further down.
- The plug-and-play interface, connecting through a single-mode fiber patch cable, lets the same rack be tested against different beam-stabilization terminals, so results transfer across link infrastructures.
- The 94% uptime over 15 hours, with the turbulence dependence quantified through $C_n^2$, indicates the system can collect the long averages needed for chronometric leveling at the centimeter level.
Reading between the lines
- The paper leaves implicit that the $2.0 \times 10^{-19}$ figure is a floor for the instrument under folded-path conditions; whether it transfers to a separated point-to-point link depends on the unmeasured assumption that shot noise and AM-PM conversion still dominate once the delay-noise term is added. That assumption is testable by comparing the folded result with the calculated delay limit.
- A testable extension is to quantify the AM-PM coefficients of individual receiver stages (limiting amplifier, mixer, counter input) and then verify that the predicted modified-Allan bump around 0.1 s disappears when those coefficients are reduced; the paper identifies the effect but not the specific dominant component.
- For future links of tens of kilometers, the cubic scaling of residual noise with link length cited from the delay-limit model means a folded testbed no longer bounds point-to-point performance; a folded link with an inserted fiber delay line could isolate the electronic floor from the atmospheric term without building a second terminal.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper presents a portable, rack-integrated two-way free-space optical frequency transfer characterization system based on a single laser split into two paths with different AOM shifts, and a fully folded 3.4 km intra-city free-space link formed by a retroreflector at the remote terminal. The key reported result is a fractional frequency instability of 2.0e-19 at 10 s averaging time (best run, measurement 1), with an uptime of 94% over 15 hours using a defined validity threshold. The authors analyze the residual non-reciprocity using Eq. (2) and attribute short-term instability to shot noise (via a parameter-free SNR estimate) and long-term instability to AM-PM conversion, calibrating the phase-versus-power response in the laboratory and applying it to measured amplitude fluctuations. The paper concludes that the system is suitable for optical clock comparisons over free space.
Significance. If the result is taken as a characterization of a technical noise floor, the paper provides a valuable, cleanly measured demonstration of how low the co-located residual noise of a two-way free-space link characterization system can be: the direct beat-note measurements, the short-fiber noise-floor control, and the parameter-free shot-noise estimate are all strengths. The measured 2.0e-19 at 10 s, even if it is the best of several runs, is a real and impressive experimental result. However, the significance for actual optical clock comparisons is conditional: because the fully folded geometry cancels delay-induced non-reciprocity exactly, the headline number does not directly measure the atmospheric non-reciprocity that a separated point-to-point link would contribute. The paper's extrapolation to that case rests on untested assumptions about the representativeness of the folded-path free-running noise and the common-mode nature of technical noise in two independent terminals. The AM-PM analysis is a plausible in-sample explanation but is not an independent validation. These limitations are acknowledged partially in Sec.
major comments (4)
- [Abstract and Sec. 3 (Fig. 3a)] The abstract's headline claim of 2.0e-19 at 10 s is the best of six runs; the paper's own Sec. 3 states that the 10-s MDEV across the six longest runs varies between 2.0e-19 and 1.1e-18 (Fig. 3a). Presenting the best run without qualification in the abstract overstates the achieved performance. The abstract should report the range or explicitly label the value as the best continuous measurement.
- [Sec. 2, 'fully folded free-space link' paragraph; Eq. (2)] In the fully folded geometry, the forward and backward beams traverse the same air path, so the delay-induced non-reciprocity terms in Eq. (2) cancel exactly, as the paper states. The measured 2.0e-19 is therefore a co-located technical noise floor (shot noise, AM-PM, and residual frequency-dependent phase shifts), not the atmospheric non-reciprocity that a separated point-to-point two-way link would contribute to a clock comparison. The Sec. 3 estimate of the delay limit for an unfolded link does not close this gap: it assumes the free-running noise recorded on the folded path and the co-located receiver noise are representative of two one-way paths measured by independent terminals. The paper should either provide a point-to-point test or explicitly frame the result as a technical-floor characterization, and temper the abstract and conclusion accordingly.
- [Sec. 3, Fig. 4(b) and Fig. 5] The AM-PM explanation of the long-term instability is supported only by an in-sample consistency check: the phase-versus-power curve is fitted to laboratory data and then applied to the amplitude fluctuations of measurement 1 to reproduce the same run's MDEV. This does not independently validate the AM-PM mechanism, and the same method cannot be assumed to transfer to a point-to-point configuration where the two directions experience different amplitude statistics, as the paper itself notes in Sec. 4 that only 'adaptations' are 'straightforward'. The manuscript should acknowledge that the agreement is consistent with, but does not uniquely prove, the AM-PM mechanism.
- [Sec. 4, Conclusion] The conclusion that 'adaptations for minimizing interferometric noise contributions in genuine point-to-point two-way free-space links ... are straightforward to implement' is not supported by any measurement in the manuscript. The system was tested only on a fully folded link and a short-fiber noise floor; no genuine point-to-point two-way free-space measurement is reported. This claim should be softened or supported by additional data, and the sentence should not imply that the reported stability directly characterizes a point-to-point comparison.
minor comments (5)
- [Sec. 2, first paragraph] The paper says 'The entire setup is interconnected using polarization-maintaining (PM) fiber' but later describes a 5 m SM fiber connection to the optical terminal and a polarization controller at the PM-to-SM transition; please clarify which parts of the setup are PM and which are SM.
- [Sec. 3, Fig. 6] The uptime is defined with a threshold of ±5e-17 for invalid data; the paper states the rationale for this threshold, but it would improve clarity to report the sensitivity of the 94% uptime figure to the threshold choice, for example by giving uptime for a range of thresholds.
- [Appendix D, Eq. (14)] The fitted coefficients a_L^Φ, b_L^Φ, c_L^Φ, d_L^Φ, a_H^Φ, b_H^Φ, and c_H^Φ are not listed; providing these numerical coefficients would aid reproducibility of the AM-PM calculation.
- [Sec. 3, near Fig. 4(b)] The phrase 'zero compatible operation' should be reworded to 'compatible with zero' for clarity.
- [Abstract and Sec. 3, Fig. 3(b)] The abstract's uptime claim of 94% over 15 hours depends on the validity threshold defined in Sec. 3; adding a qualifier such as 'using the ±5e-17 threshold' in the abstract or a footnote would avoid ambiguity.
Circularity Check
No circularity: the headline instability is a direct measurement of the folded-link residual, and the noise models are independent cross-checks, not fits to the target.
full rationale
The central result is an experimental measurement, not a derivation. The residual two-way noise is obtained directly from recorded beat-note counts via Eq. (2), with common-mode interferometric and laser-phase noise canceled. The fully folded geometry deliberately cancels delay-induced non-reciprocity, so the quoted 2.0e-19 is explicitly the residual technical noise of that configuration; this is a scope limitation for point-to-point extrapolation, not a circular reduction. The AM-PM explanation is not circular: the phase-versus-RF-power curve was fit to separate controlled laboratory data (Fig. 5(a)) and then applied to the independently recorded amplitude fluctuations from Measurement 1 to generate a predicted phase noise MDEV (purple curve), which is compared with the measured MDEV rather than used to produce it. Similarly, the shot-noise floor is calculated from standard heterodyne formulas using the measured RF beat power and local-oscillator power, and then compared with the measured MDEV. The red-dashed delay-limit curve for an unfolded link is an external estimate using the measured free-running noise, not a fit to the observed residual. Self-citations [26] and [36] are used only for auxiliary details (partially folded configuration and straightforward adaptations) and are not load-bearing for the headline instability. No step in the derivation chain is defined in terms of its conclusion.
Assumptions & free parameters
free parameters (2)
- AM-PM phase calibration coefficients (piecewise polynomial/exponential) =
not reported numerically in the paper
- Uptime validity threshold =
+/-5x10^-17 fractional
assumptions (4)
- domain assumption All optical paths not explicitly listed are either common-mode to both beat notes or part of the two-way compensated path (Appendix A, Eqs. (8)-(9)).
- domain assumption In the fully folded link, the forward and backward beams experience the same delay-induced phase shift, so the corresponding non-reciprocity cancels exactly.
- domain assumption The photodetection is shot-noise limited, with local oscillator shot noise dominating other detector noise (Sec. 3, Eqs. (5)-(7)).
- standard math Doppler-induced non-reciprocity from the AOM frequency offset is negligible at the 5e-8 level relative to the free-running Doppler shift (Appendix C, Eq. (12)).
Cite this review
Pith. "Pith review of A plug-and-play solution for characterizing two-way optical frequency transfer over free-space." pith.science (2026). https://pith.science/paper/2JYKTPBA
@misc{pith2026250202161,
author = {Pith},
title = {Pith review of: A plug-and-play solution for characterizing two-way optical frequency transfer over free-space},
year = {2026},
howpublished = {\url{https://pith.science/paper/2JYKTPBA}},
note = {Machine review of arXiv:2502.02161}
}
abstract
Optical clock networks connected by phase-coherent links offer significant potential for advancing fundamental research and diverse scientific applications. Free-space optical frequency transfer extends fiber-based connectivity to remote areas and holds the potential for global coverage via satellite links. Here we present a compact and robust portable, rack-integrated two-way free-space link characterization system. Equipped with plug-and-play capabilities, the system enables straightforward interfacing with various optical systems and facilitates quick deployment for field experiments. In this work, we achieve a fractional frequency instability of $2.0 \times 10^{-19}$ for an averaging time of 10 s over a 3.4 km horizontal fully folded intra-city free-space link. Moreover, the system maintains an uptime of $94\%$ over 15 hours, illustrating its reliability and effectiveness for high-precision optical frequency comparisons over free-space.
Figures
Figures from the paper (4 more)
Reference graph
Works this paper leans on
-
[1]
Cryogenic optical lattice clocks,
I. Ushijima, M. Takamoto, M. Das,et al., “Cryogenic optical lattice clocks,” Nat. Photonics9, 185–189 (2015)
work page 2015
-
[2]
Single-ion atomic clock with3× 10−18 systematic uncertainty,
N. Huntemann, C. Sanner, B. Lipphardt,et al., “Single-ion atomic clock with3× 10−18 systematic uncertainty,” Phys. Rev. Lett.116, 063001 (2016)
work page 2016
-
[3]
27Al+ quantum-logic clock with a systematic uncertainty below 10−18,
S. M. Brewer, J.-S. Chen, A. M. Hankin,et al., “27Al+ quantum-logic clock with a systematic uncertainty below 10−18,” Phys. Rev. Lett.123, 033201 (2019)
work page 2019
-
[4]
176Lu+ clock comparison at the10−18 level via correlation spectroscopy,
Z. Zhang, K. J. Arnold, R. Kaewuam, and M. D. Barrett, “176Lu+ clock comparison at the10−18 level via correlation spectroscopy,” Sci. Adv.9, 1971 (2023)
work page 2023
-
[5]
Clock with8× 10−19 systematic uncertainty,
A. Aeppli, K. Kim, W. Warfield,et al., “Clock with8× 10−19 systematic uncertainty,” Phys. Rev. Lett.133, 023401 (2024)
work page 2024
- [6]
-
[7]
Quantum tests of the Einstein equivalence principle with the STE–QUEST space mission,
B. Altschul, Q. G. Bailey, L. Blanchet,et al., “Quantum tests of the Einstein equivalence principle with the STE–QUEST space mission,” Adv. Space Res.55, 501–524 (2015)
work page 2015
-
[8]
Varying constants, gravitation and cosmology,
J.-P. Uzan, “Varying constants, gravitation and cosmology,” Living Rev. Relativ.14, 2 (2011)
work page 2011
Show all 36 references
-
[9]
Atomic clocks for geodesy,
T. E. Mehlstäubler, G. Grosche, C. Lisdat,et al., “Atomic clocks for geodesy,” Reports on Prog. Phys.81, 064401 (2018)
2018
-
[10]
Atomic clock performance enabling geodesy below the centimetre level,
W. F. McGrew, X. Zhang, R. J. Fasano,et al., “Atomic clock performance enabling geodesy below the centimetre level,” Nature564, 87–90 (2018)
2018
-
[11]
Space clocks and fundamental tests: The ACES experiment,
L. Cacciapuoti and C. Salomon, “Space clocks and fundamental tests: The ACES experiment,” The Eur. Phys. J. Special Top.172, 57–68 (2009)
2009
-
[12]
Deliveringthesameopticalfrequencyattwoplaces: accuratecancellation of phase noise introduced by an optical fiber or other time-varying path,
L.-S.Ma,P.Jungner,J.Ye,andJ.L.Hall,“Deliveringthesameopticalfrequencyattwoplaces: accuratecancellation of phase noise introduced by an optical fiber or other time-varying path,” Opt. Lett.19, 1777–1779 (1994)
1994
-
[13]
Remote transfer of ultrastable frequency references via fiber networks,
S. M. Foreman, K. W. Holman, D. D. Hudson,et al., “Remote transfer of ultrastable frequency references via fiber networks,” Rev. Sci. Instruments78, 021101 (2007)
2007
-
[14]
Brillouin amplification supports1× 10−20 uncertainty in optical frequency transfer over 1400 km of underground fiber,
S. M. F. Raupach, A. Koczwara, and G. Grosche, “Brillouin amplification supports1× 10−20 uncertainty in optical frequency transfer over 1400 km of underground fiber,” Phys. Rev. A92, 021801 (2015)
2015
-
[15]
Cascaded optical fiber link using the internet network for remote clocks comparison,
N. Chiodo, N. Quintin, F. Stefani,et al., “Cascaded optical fiber link using the internet network for remote clocks comparison,” Opt. Express23, 33927–33937 (2015)
2015
-
[16]
Impact of turbulence on high-precision ground-satellite frequency transfer with two-way coherent optical links,
C. Robert, J.-M. Conan, and P. Wolf, “Impact of turbulence on high-precision ground-satellite frequency transfer with two-way coherent optical links,” Phys. Rev. A93, 033860 (2016)
2016
-
[17]
Andrews and R
L. Andrews and R. Phillips,Laser beam propagation through Random Media, 2nd ed.(SPIE Press, 2005, Chap. 3)
2005
-
[18]
Optical two-way time and frequency transfer over free space,
F. R. Giorgetta, W. C. Swann, L. C. Sinclair,et al., “Optical two-way time and frequency transfer over free space,” Nat. Photonics7, 434–438 (2013)
2013
-
[19]
Free-space dissemination of time and frequency with10−18 instability over 113km,
Q. Shen, J.-Y. Guan, J.-G. Ren,et al., “Free-space dissemination of time and frequency with10−18 instability over 113km,” Nature610, 661–666 (2022)
2022
-
[20]
Quantum-limited optical time transfer for future geosynchronous links,
E. D. Caldwell, J.-D. Deschenes, J. Ellis,et al., “Quantum-limited optical time transfer for future geosynchronous links,” Nature618, 721–726 (2023)
2023
-
[21]
Free-spacetransferofcomb-rootedopticalfrequenciesoveran18kmopen-air link,
H.J.Kang,J.Yang,B.J.Chun, etal.,“Free-spacetransferofcomb-rootedopticalfrequenciesoveran18kmopen-air link,” Nat. Commun.10, 4438 (2019)
2019
-
[22]
Point-to-point stabilized optical frequency transfer with active optics,
B. P. Dix-Matthews, S. W. Schediwy, D. R. Gozzard,et al., “Point-to-point stabilized optical frequency transfer with active optics,” Nat. Commun.12, 515 (2021)
2021
-
[23]
Ultrastable free-space laser links for a global network of optical atomic clocks,
D. R. Gozzard, L. A. Howard, B. P. Dix-Matthews,et al., “Ultrastable free-space laser links for a global network of optical atomic clocks,” Phys. Rev. Lett.128, 020801 (2022)
2022
-
[24]
Free space optical link to a tethered balloon for frequency transfer and chronometric geodesy,
N. Maron, S. Fernandez, F.-X. Esnault,et al., “Free space optical link to a tethered balloon for frequency transfer and chronometric geodesy,” Opt. Express32, 4267–4276 (2024)
2024
-
[25]
Frequency transfer via a two-way optical phase comparison on a multiplexed fiber network,
C. E. Calosso, E. Bertacco, D. Calonico,et al., “Frequency transfer via a two-way optical phase comparison on a multiplexed fiber network,” Opt. Lett.39, 1177–1180 (2014)
2014
-
[26]
A two-way free-space link for optical frequency comparisons,
J. Ji, A. Kuhl, J. Kronjäger, and S. Koke, “A two-way free-space link for optical frequency comparisons,” in2023 JointConferenceoftheEuropeanFrequencyandTimeForumandIEEEInternationalFrequencyControlSymposium, (2023, paper 7050)
2023
-
[27]
On the relation between uncertainties of weighted frequency averages and the various types of allan deviations,
E. Benkler, C. Lisdat, and U. Sterr, “On the relation between uncertainties of weighted frequency averages and the various types of allan deviations,” Metrologia52, 565 (2015)
2015
-
[28]
High performance optical free-space links for quantum communications,
M. Goy, R. Berlich, A. Kržič,et al., “High performance optical free-space links for quantum communications,” in International Conference on Space Optics — ICSO 2020,vol. 11852 B. Cugny, Z. Sodnik, and N. Karafolas, eds., International Society for Optics and Photonics (SPIE, 20...
2020
-
[29]
Coherent transfer of an optical carrier over 251 km,
N. R. Newbury, P. A. Williams, and W. C. Swann, “Coherent transfer of an optical carrier over 251 km,” Opt. Lett. 32, 3056–3058 (2007)
2007
-
[30]
Cascaded multiplexed optical link on a telecommunication network for frequency dissemination,
O. Lopez, A. Haboucha, F. Kéfélian,et al., “Cascaded multiplexed optical link on a telecommunication network for frequency dissemination,” Opt. Express18, 16849–16857 (2010)
2010
-
[31]
Tackling the limits of optical fiber links,
F. Stefani, O. Lopez, A. Bercy,et al., “Tackling the limits of optical fiber links,” J. Opt. Soc. Am. B32, 787–797 (2015)
2015
-
[32]
R. H. Kingston,Coherent or Heterodyne Detection(Springer, 1978, pp. 24–38)
1978
-
[33]
The Enrico’s chart for phase noise and two-sample variances,
E. Rubiola, “The Enrico’s chart for phase noise and two-sample variances,” (2023). https://doi.org/10.5281/zenodo.4399218
2023 doi
-
[34]
Long-distance chronometric leveling with a portable optical clock,
J. Grotti, I. Nosske, S. Koller,et al., “Long-distance chronometric leveling with a portable optical clock,” Phys. Rev. Appl. 21, L061001 (2024)
2024
-
[35]
Opticalfrequencydistributionusinglaserrepeaterstationswithplanarlightwave circuits,
T.Akatsuka,T.Goh,H.Imai, etal.,“Opticalfrequencydistributionusinglaserrepeaterstationswithplanarlightwave circuits,” Opt. Express28, 9186–9197 (2020)
2020
-
[36]
Field-deployableinterferometricfibrelinkterminalswithverylowtemperaturesensitivity,
A.KuhlandJ.Kronjäger,“Field-deployableinterferometricfibrelinkterminalswithverylowtemperaturesensitivity,” J. Physics: Conf. Ser.2889, 012029 (2024)
2024
Reviewed August 9, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.