Multiscale phase dynamics and $2\pi$ phase kinks in injection-locked optoelectronic oscillators with large delay

Abhijit Banerjee; Trevor J. Hall

arxiv: 2604.18499 · v1 · submitted 2026-04-20 · ⚛️ physics.optics

Multiscale phase dynamics and 2π phase kinks in injection-locked optoelectronic oscillators with large delay

Abhijit Banerjee , Trevor J. Hall This is my paper

Pith reviewed 2026-05-10 04:00 UTC · model grok-4.3

classification ⚛️ physics.optics

keywords injection lockingoptoelectronic oscillatorsphase kinksdelay differential equationsAdler equationsmultiscale analysisphase dynamicsfrequency locking

0 comments

The pith

Large-delay injection-locked optoelectronic oscillators spontaneously form persistent 2π phase kinks that enable frequency locking without phase locking.

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

The paper develops a multiscale theoretical framework to explain the origin and stability of spontaneous 2π phase kinks in injection-locked optoelectronic oscillators with large delays. Starting from a complex-envelope delay differential equation that includes gain saturation and bandpass filter dynamics, the authors derive a phase-only model and apply a two-timescale reduction to obtain a continuum of Adler equations for the phase difference as a function of round-trip time. Analytical solutions for weak injection show how initially smooth phase profiles sharpen into localized kinks, with p kinks appearing per round trip near the pth adjacent mode. Full-model simulations confirm the sharpening mechanism but demonstrate that resonator-induced amplitude excursions can erase kinks when the trajectory does not encircle the origin. This matters because it extends classical single-mode locking theory to large-delay regimes and identifies a distinct operating state of frequency locking without phase locking.

Core claim

A two-timescale reduction applied to the phase dynamics of large-delay IL-OEOs yields a continuum of Adler equations that govern the slow evolution of the phase difference between the injected signal and the oscillator. For weak injection and small detuning, the equations predict that smooth phase profiles evolve into p localized 2π phase kinks per round trip when tuned near the pth adjacent mode. Time-domain simulations of the full complex-envelope model validate the kink formation process while revealing that RF resonator dynamics control persistence: amplitude excursions associated with steep phase gradients can erase the kinks and restore conventional phase locking.

What carries the argument

A continuum of Adler equations obtained via two-timescale reduction of the phase difference between injected signal and oscillator as a function of round-trip time, which drives the sharpening of phase profiles into localized 2π kinks.

If this is right

Initially smooth phase profiles sharpen into localized 2π phase kinks under weak injection and small detuning.
p kinks appear per round trip when the injection is tuned near the pth adjacent mode.
The reduced phase-only model correctly predicts kink formation but is limited by amplitude variations.
Resonator-induced amplitude excursions can erase kinks when the trajectory fails to encircle the origin.
Frequency locking without phase locking occurs in the regime where kinks persist.

Where Pith is reading between the lines

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

The same multiscale reduction may apply to other classes of large-delay oscillators in optics or electronics.
Varying the RF bandpass filter bandwidth offers a direct test of the boundary between phase-locked and frequency-locked states.
Persistent kinks could be exploited to store or manipulate phase information across multiple round trips.
Adding weak noise to the reduced model would clarify whether kinks remain stable against small perturbations.

Load-bearing premise

The oscillation amplitude remains nearly constant so that a phase-only description accurately captures the dynamics.

What would settle it

A simulation or experiment in which a steep phase gradient causes the complex-envelope trajectory to fail to encircle the origin, after which the predicted kinks disappear and conventional locking is restored.

read the original abstract

Injection locking of optoelectronic oscillators (OEOs) with large delay gives rise to phase dynamics that lie beyond the scope of classical single mode locking theory, including the spontaneous formation of persistent $2\pi$ phase kinks. In this work, a multiscale theoretical framework is developed that explains the origin, structure, and stability of these phase slip phenomena in injection locked (IL) OEOs operating in large-delay regime. Starting from a complex envelope delay differential equation that explicitly incorporates hard-limiting gain saturation and RF BPF dynamics, a reduced phase-only description valid for nearly constant oscillation amplitude is derived. Exploiting the separation between fast round-trip dynamics and slow inter-round-trip evolution, a two-timescale reduction yields a continuum of Adler equations governing phase difference between injected signal and oscillator as a function of round-trip time. Analytical solutions obtained for weak injection and small detuning show how initially smooth phase profiles sharpen into localized $2\pi$ phase kinks, with p kinks appearing per round trip when the injection is tuned near the pth adjacent mode. Time-domain simulations of full complex-envelope model validate the predicted phase kink formation mechanism and reveal essential role of RF resonator dynamics in determining their persistence. While the reduced phase-only model correctly predicts kink sharpening, resonator-induced amplitude excursions associated with steep phase gradients can erase the kinks when the complex-envelope trajectory fails to encircle the origin, restoring conventional phase locking. These results provide a unified physical interpretation of $2\pi$phase kinks in IL-OEOs and delineate the limits of phase-only models in the presence of large, fast phase transients, identifying a regime of frequency locking without phase locking in large delay oscillators.

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.

Desk Editor's Note private letter to a colleague

The paper gives a usable multiscale phase model for 2π kinks in large-delay IL-OEOs while marking where the constant-amplitude assumption stops working.

read the letter

The main point is a multiscale reduction that maps the large-delay injection-locked OEO onto a continuum of Adler equations. This predicts the formation of p localized 2π phase kinks per round trip and a regime of frequency locking without phase locking. The paper does a clean job deriving the phase-only model from the full complex-envelope DDE that includes gain saturation and the RF filter. It separates the fast round-trip time from the slow inter-round-trip evolution and solves the resulting equations analytically for weak injection. The time-domain simulations then show how the kinks actually behave and when the RF resonator wipes them out through amplitude changes. This is a step past classical single-mode locking theory for the large-delay regime. The authors are straightforward about the assumption of nearly constant amplitude and about the cases where it fails. The soft spot is the one the stress test highlights. The phase-only description assumes amplitude stays steady, yet the steep gradients at the kinks drive amplitude excursions that can prevent the envelope from encircling the origin and erase the kinks. The reduced model therefore explains the sharpening but leaves the persistence to the full simulations. That is not a fatal flaw, just a clear boundary on what the analytic part can say. The citation pattern looks standard and there is no sign of circular fitting. This work is for people who study optoelectronic oscillators and microwave photonics. A reader in that area gets a practical framework plus an honest map of where phase-only thinking stops working. It deserves a serious referee because the derivation is traceable, the simulations provide direct checks, and the subfield can use the insight into kink stability. I would send it for peer review. The contribution is focused and the evidence supports the claims within the stated limits.

Referee Report

1 major / 2 minor

Summary. The paper develops a multiscale theoretical framework starting from a complex-envelope delay differential equation (DDE) for injection-locked optoelectronic oscillators (OEOs) that includes hard-limiting gain saturation and RF bandpass filter dynamics. It derives a reduced phase-only model under the assumption of nearly constant amplitude, performs a two-timescale reduction to a continuum of Adler equations for the phase difference as a function of round-trip time, and obtains analytical solutions for weak injection and small detuning that predict sharpening of phase profiles into localized 2π kinks (p kinks per round trip near the pth mode). Time-domain simulations of the full model are used to validate kink formation while also showing that resonator-induced amplitude excursions can erase kinks when the complex-envelope trajectory fails to encircle the origin, thereby identifying a regime of frequency locking without phase locking.

Significance. If the central claims hold, the work offers a useful physical interpretation of 2π phase kinks and phase slips in large-delay IL-OEOs that extends classical single-mode Adler locking theory. The explicit delineation of the limits of the phase-only reduction (via amplitude excursions) is a strength, as is the combination of analytical solutions with full-model simulations. This could inform design of OEOs for microwave photonics applications where delay-induced phase dynamics matter.

major comments (1)

[Abstract and the section presenting the two-timescale reduction to Adler equations] The phase-only reduction (derived from the complex-envelope DDE under the nearly-constant-amplitude assumption) is load-bearing for the predicted kink sharpening and stability, yet the abstract and simulation discussion explicitly state that steep phase gradients induce amplitude excursions that erase the kinks and restore conventional locking. This creates an internal tension: the reduced model is used to explain the origin and structure of the very features whose persistence requires the approximation to fail. A quantitative bound on amplitude variation (e.g., via the injection strength or detuning parameters) or a direct comparison of phase profiles from the reduced vs. full model is needed to establish the regime of validity.

minor comments (2)

[Section on multiscale reduction] The notation for the continuum limit of the Adler equations and the definition of the slow time scale could be made more explicit with numbered equations to aid reproducibility.
[Simulation results section] The manuscript would benefit from a brief table or plot quantifying the amplitude excursion magnitude (e.g., |A| variation) across the parameter space where kinks are predicted vs. erased.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the positive assessment of our work and for the constructive major comment. We agree that the regime of validity of the phase-only reduction requires clearer delineation and will revise the manuscript accordingly.

read point-by-point responses

Referee: [Abstract and the section presenting the two-timescale reduction to Adler equations] The phase-only reduction (derived from the complex-envelope DDE under the nearly-constant-amplitude assumption) is load-bearing for the predicted kink sharpening and stability, yet the abstract and simulation discussion explicitly state that steep phase gradients induce amplitude excursions that erase the kinks and restore conventional locking. This creates an internal tension: the reduced model is used to explain the origin and structure of the very features whose persistence requires the approximation to fail. A quantitative bound on amplitude variation (e.g., via the injection strength or detuning parameters) or a direct comparison of phase profiles from the reduced vs. full model is needed to establish the regime of validity.

Authors: We acknowledge the tension identified by the referee. The phase-only reduction is intended to isolate the mechanism by which smooth phase profiles sharpen into 2π kinks under the nearly-constant-amplitude assumption (valid for weak injection and small detuning). The manuscript already notes that this assumption breaks down once gradients steepen, allowing resonator-induced amplitude excursions to erase the kinks. To resolve the concern, the revised manuscript will add (i) an explicit quantitative bound on amplitude variation, obtained by estimating the maximum deviation of |A| from its mean value as a function of injection strength and detuning in the two-timescale analysis, and (ii) a direct side-by-side comparison of phase profiles extracted from the reduced continuum of Adler equations versus the full complex-envelope DDE simulations, for parameter sets both inside and outside the regime where kinks persist. These additions will be placed in the section on the two-timescale reduction and supported by a new figure. revision: yes

Circularity Check

0 steps flagged

No significant circularity; standard reductions with explicit validity limits

full rationale

The derivation begins from the complex-envelope DDE, invokes the explicit assumption of nearly constant amplitude to obtain a phase-only model, then applies a standard two-timescale separation to reach a continuum of Adler equations. Analytical solutions for kink sharpening are obtained under weak injection and small detuning; these are cross-validated by direct time-domain integration of the full complex-envelope model rather than by construction. The abstract itself states the breakdown condition (amplitude excursions erasing kinks when the trajectory fails to encircle the origin), so the claimed regime of frequency locking without phase locking is presented as a qualified outcome of the full model, not a tautological restatement of the reduced equations. No load-bearing self-citations, fitted parameters renamed as predictions, or ansatz smuggling appear in the derivation chain.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 0 invented entities

The central claim rests on assumptions and parameters extracted from the abstract; the full paper may specify numerical values or additional details.

free parameters (2)

injection strength
Analytical solutions obtained for weak injection
detuning
Small detuning assumed in solutions

axioms (2)

domain assumption Nearly constant oscillation amplitude
Assumed to derive the phase-only description
domain assumption Separation between fast round-trip dynamics and slow inter-round-trip evolution
Exploited for the two-timescale reduction

pith-pipeline@v0.9.0 · 5610 in / 1477 out tokens · 68345 ms · 2026-05-10T04:00:10.226644+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

11 extracted references · 11 canonical work pages

[1]

Banerjee, T

A. Banerjee, T. J. Hall, ‘Simulation of optoelectronic oscillator injection locking, pulling and spiking phenomena’ , Scientific Reports, 14, 4332, (2024)

work page 2024
[2]

D. P . Rosin, ‘Pulse train solutions in a time-delayed opto-electronic oscillator’ , Master’s thesis, Technische Universität Berlin, (2011)

work page 2011
[3]

H. Tian, L. Zhang, Z. Zeng, Y . Wu, Z. Fu, W. Lyu, Y . Zhang, Z. Zhang, S. Zhang, H. Li, H. Y . Liu, ‘Externally triggered spiking and spiking cluster oscillation in broadband optoelectronic oscillator’ , J. Lightwave Technology, 41(1), 48-58, (2022). Injection locked OEO multiscale phase dynamics 25

work page 2022
[4]

Diakonov, M

A. Diakonov, M. Horowitz, ‘Generation of ultra-low jitter radio frequency phase pulses by a phase-locked oscillator’ , Opt. Lett., 46(19),5047-5050, (2021)

work page 2021
[5]

Smulakovsky, A

V . Smulakovsky, A. Diakonov, A. Katzenlson, M. Horowitz, ‘Temporal locking of pulses in injection locked oscillators, ’ Scientific Reports, 15, 5602, (2025)

work page 2025
[6]

E. C. Levy, M. Horowitz, C. R. Menyuk, ‘Modeling optoelectronic oscillators’, J. Opt. Soc. Am. B, 26(1),148-159, (2009)

work page 2009
[7]

A simple model of feedback oscillator noise spectrum,

D. B. Leeson, “A simple model of feedback oscillator noise spectrum, ” Proc. IEEE, 329- 330, (1966)

work page 1966
[8]

D. B. Leeson, ‘Oscillator phase noise: a 50-years review’, IEEE Trans. Ultrasonics, Ferroelectrics & Frequency Control, 63(8), 1208-1225, (2016)

work page 2016
[9]

L. J. Paciorek, ‘Injection locking of oscillators’, Proc. IEEE, 53(11), 1723-1727, (1965)

work page 1965
[10]

A study of locking phenomena in oscillators,

R. Adler, “A study of locking phenomena in oscillators, ” Proc. IRE, 351-357, (1946)

work page 1946
[11]

R. M. Corless, G. H. Gonnet, D. E. G. Hare, D. J. Jeffrey, D. E. Knuth, ‘On the Lambert W function’ , Advances in Computational Mathematics 5(1) 329-359 (1996). Injection locked OEO multiscale phase dynamics 26 Supplementary Material Linear analysis of the free oscillator The complex envelope DDE model (main text Equation 10) is linearized by treating the...

work page 1996

[1] [1]

Banerjee, T

A. Banerjee, T. J. Hall, ‘Simulation of optoelectronic oscillator injection locking, pulling and spiking phenomena’ , Scientific Reports, 14, 4332, (2024)

work page 2024

[2] [2]

D. P . Rosin, ‘Pulse train solutions in a time-delayed opto-electronic oscillator’ , Master’s thesis, Technische Universität Berlin, (2011)

work page 2011

[3] [3]

H. Tian, L. Zhang, Z. Zeng, Y . Wu, Z. Fu, W. Lyu, Y . Zhang, Z. Zhang, S. Zhang, H. Li, H. Y . Liu, ‘Externally triggered spiking and spiking cluster oscillation in broadband optoelectronic oscillator’ , J. Lightwave Technology, 41(1), 48-58, (2022). Injection locked OEO multiscale phase dynamics 25

work page 2022

[4] [4]

Diakonov, M

A. Diakonov, M. Horowitz, ‘Generation of ultra-low jitter radio frequency phase pulses by a phase-locked oscillator’ , Opt. Lett., 46(19),5047-5050, (2021)

work page 2021

[5] [5]

Smulakovsky, A

V . Smulakovsky, A. Diakonov, A. Katzenlson, M. Horowitz, ‘Temporal locking of pulses in injection locked oscillators, ’ Scientific Reports, 15, 5602, (2025)

work page 2025

[6] [6]

E. C. Levy, M. Horowitz, C. R. Menyuk, ‘Modeling optoelectronic oscillators’, J. Opt. Soc. Am. B, 26(1),148-159, (2009)

work page 2009

[7] [7]

A simple model of feedback oscillator noise spectrum,

D. B. Leeson, “A simple model of feedback oscillator noise spectrum, ” Proc. IEEE, 329- 330, (1966)

work page 1966

[8] [8]

D. B. Leeson, ‘Oscillator phase noise: a 50-years review’, IEEE Trans. Ultrasonics, Ferroelectrics & Frequency Control, 63(8), 1208-1225, (2016)

work page 2016

[9] [9]

L. J. Paciorek, ‘Injection locking of oscillators’, Proc. IEEE, 53(11), 1723-1727, (1965)

work page 1965

[10] [10]

A study of locking phenomena in oscillators,

R. Adler, “A study of locking phenomena in oscillators, ” Proc. IRE, 351-357, (1946)

work page 1946

[11] [11]

R. M. Corless, G. H. Gonnet, D. E. G. Hare, D. J. Jeffrey, D. E. Knuth, ‘On the Lambert W function’ , Advances in Computational Mathematics 5(1) 329-359 (1996). Injection locked OEO multiscale phase dynamics 26 Supplementary Material Linear analysis of the free oscillator The complex envelope DDE model (main text Equation 10) is linearized by treating the...

work page 1996