REVIEW 4 major objections 5 minor 1 cited by
Phase Biasing System for Optical Gyroscope Using Passive Non-Reciprocal Polarization Techniques
T0 review · 4 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash
Pith's one-line read This paper seeks to establish that a passive non-reciprocal polarization phase shifter can replace active modulation in a fiber-optic gyroscope, producing two simultaneous quadrature signals and cutting simulated angular random walk by up…
desk verdict Interesting passive-biasing idea for IFOGs, but the 40x ARW claim is an artifact of a noiseless simulation. 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 Non-Reciprocal Polarization-Dependent Phase Shifter (NRPPS): a passive chain consisting of a polarizing beam splitter, a non-reciprocal element that rotates polarization by +45 degrees on each pass, an eighth-wave plate retarder, and a mirror. Because the clockwise beam crosses the retarder with its fast axis matched while the counter-clockwise beam crosses with its slow axis matched, the round trip adds phase $2\phi_r$ to the counter-clockwise beam and no net phase to the clockwise beam; with $\phi_r = \pi/4$ this is exactly $\pi/2$. Two output ports then supply the quadrature signals at relative phases $\pi + 2\phi_r$ and $2\phi_r$, which become $3\pi/2$ and $\pi/2$ when the retarder is an eighth-wave plate. The paper derives this with Jones matrices for each polarization state.
What would settle it
Measure the overlapping Allan variance of the difference of the two detector outputs in a stationary NRPPS-IFOG while attenuating the optical power; if the noise floor scales with the individual detector shot-noise slopes instead of vanishing, the simulated 10-40x ARW improvement is not attainable.
Extended reading notes
Core claim
The central discovery is that passive quadrature biasing can be obtained in an IFOG by inserting an NRPPS into the beam path. The non-reciprocal polarization rotation makes the clockwise and counter-clockwise beams experience different retarder alignments, so one round trip accumulates a phase $2\phi_r$ while the other accumulates none. Setting the retarder as an eighth-wave plate, $\phi_r = \pi/4$, places the two interference outputs at the quadrature points $\pi/2$ and $3\pi/2$, giving complementary readouts. The paper claims this is the first IFOG configuration to operate at two quadrature points simultaneously without active modulation, and that the complementary readouts support common-mode noise cancellation that reduces angular random walk by factors of about 10, 30, and 40 for 200 m, 1000 m, and 2000 m fiber coils.
Load-bearing premise
The load-bearing premise is that the two photodetector outputs share perfectly correlated, time-matched noise so that subtracting them removes all noise; the paper simulates this ideal cancellation and does not model detector shot noise, thermal noise, or polarization drift, which are not common-mode in a real gyroscope.
Editorial extensions
If this is right
- Replacing the active modulator with the NRPPS removes the drive electronics, power supply, and thermal stabilization that conventional MIOC or piezoelectric biasing requires.
- Because the two readouts sit at $\pi/2$ and $3\pi/2$, the gyroscope can produce one rotation-rate data point per coil cycle rather than one per two cycles, giving continuous readout.
- The coil length no longer has to respect modulation timing constraints, so designers can choose longer fiber coils for higher sensitivity without changing the biasing scheme.
- If the simulated common-mode noise cancellation is real, angular random walk falls by roughly 10x, 30x, and 40x for 200 m, 1000 m, and 2000 m coils compared with a conventional IFOG.
- The passive bias principle generalizes to other interferometric gyroscopes and Sagnac sensors that need a fixed phase bias between counter-propagating beams.
Reading between the lines
- The paper's own caveat that a perfect temporal match makes the system noiseless implies the 40x gain is an idealization: real detector shot noise and thermal noise are not common-mode, so the gain in hardware is likely smaller.
- A closed-loop variant is implicit in the design: placing a piezoelectric actuator behind the NRPPS mirror, as the paper mentions for future work, would allow feedback to hold the quadrature bias point and extend dynamic range.
- The two-quadrature readout resembles balanced detection, so the scheme could be adapted to other Sagnac interferometers, including chip-scale or resonant gyroscopes, wherever a passive $\pi/2$ bias is needed.
- A direct experimental check is to measure the Allan variance of the PD2 minus PD3 signal with the input beam blocked; if the residual noise is set by the detector floors, the ARW improvement will be bounded by detector performance.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript proposes a passive phase-biasing scheme for an interferometric fiber optic gyroscope (IFOG) based on a Non-Reciprocal Polarization-Dependent Phase Shifter (NRPPS). The central claim is that the NRPPS introduces a fixed π/2 phase shift between counter-propagating beams without active modulation, enabling simultaneous operation at two quadrature points (π/2 and 3π/2) and providing natural noise suppression. The authors present a qualitative optical architecture, a Jones calculus model of the NRPPS element, and MATLAB simulations comparing the angular random walk (ARW) of their NRPPS-IFOG with a conventional IFOG for 200 m, 1000 m, and 2000 m fiber coils. They report ARW improvements of roughly 10×, 30×, and 40×, respectively, and argue that passive biasing reduces power consumption, complexity, and long-term drift.
Significance. If the proposed scheme worked as described, it would be a useful contribution to IFOG design: passive biasing at two quadrature points would eliminate active modulators and their associated control electronics, with potential benefits in power, robustness, and miniaturization. The manuscript correctly identifies a real problem and engages with relevant prior work on passive biasing, and the explicit attempt to provide a Jones calculus model and a parametric simulation study is commendable. However, the significance is not established by the present manuscript. The Jones calculus contains a load-bearing error (the PBS is set to the identity matrix, which removes the polarization-dependent routing the entire architecture relies on), the phase budget that yields the quadrature points omits the necessary explanation of the π shift at the coupler, and the quantitative ARW improvement is derived from a simulation with no stochastic noise model. The claimed 40× improvement is therefore unsupported by the evidence presented.
major comments (4)
- [Section 2, Eq. (11) and Eqs. (17)-(19)] Setting the PBS Jones matrix J_PBS to the 2×2 identity removes the polarization-dependent routing that the entire NRPPS architecture relies on. The text explains that the CW beam passes straight through the PBS while the CCW beam is reflected, and that after the double pass the rotated polarization is rerouted to a different output port. With J_PBS=I, Eqs. (18)-(19) describe a single spatial path with no polarization-dependent separation; the Jcoll2(90°) matrix at the output is then applied without physical justification, and the claimed result E_CW_out=H does not follow from the stated matrices. A valid model requires separate transmission and reflection matrices for the two orthogonal polarizations, or a 4-port scattering description of the PBS.
- [Section 2, Eqs. (1)-(8) and Eqs. (20)-(24)] The phase budget that leads to the claimed quadrature operation is inconsistent. The Jones calculation in Eq. (22) yields a relative phase of 2ϕr between the two counter-propagating outputs, but Eq. (1) assigns a relative phase of π+2ϕr to the PD2 interference without deriving the π term from a coupler scattering matrix or from an explicit path difference. Since the claimed simultaneous operation at π/2 and 3π/2 depends directly on this π, the reader cannot verify the core bias mechanism from the presented equations. In addition, Eq. (15) defines the retarder with diag(e^{iϕr}, e^{-iϕr}); for a true eighth-wave plate the retardance should be π/4, implying ϕr=π/8, not ϕr=π/4 as stated. The value ϕr=π/4 corresponds to a quarter-wave retardance, which changes the output phases and needs to be reconciled.
- [Simulation section, Tables 1-2] The reported ARW improvement is an artifact of a noiseless deterministic model. The text states that 'perfect temporal match makes the system noiseless' and introduces only a 3 m fiber delay between PD2 and PD3; no stochastic noise source is modeled, including detector shot noise, thermal noise, relative intensity noise, polarization drift, or thermal phase noise. Subtracting or combining two deterministic signals that are identical apart from a time delay cannot produce a physically meaningful Allan variance, and the reported values such as 0.00002 °/√hr for the 2000 m coil are orders of magnitude below any realistic shot-noise-limited estimate for a 100 mW source at 1550 nm with typical losses. The comparison in Table 1 is therefore not a valid basis for the claimed 10×-40× improvement in ARW.
- [Simulation section, paragraph beginning 'In the simulation, sensitivity was calculated...'] The input rotation rate is stated as 'Earth's rotation (15°/s)', but the Earth's rotation rate is 15°/hour, not 15°/s. This factor-3600 error enters the Sagnac phase and all rotation-rate-dependent quantities used in the Allan variance analysis. The simulation parameters must be corrected and the ARW comparison re-run before the quantitative claims can be assessed.
minor comments (5)
- [Section 2, Eq. (20)] E_CCW_in is set equal to H, although the text describes the CCW beam as vertically polarized when it enters the PBS; this is inconsistent with the stated routing and should be corrected.
- [Section 2, Eqs. (22)-(24)] The output of the CCW-path calculation is labeled E_CW_out in Eqs. (22)-(24); these labels should be E_CCW_out to avoid confusion with the CW-path result in Eq. (19).
- [Throughout] The text repeatedly uses '8-wave plate' and '8-wave retarder' where 'eighth-wave plate' is meant; given Eq. (15), the value ϕr=π/4 corresponds to a quarter-wave retardance, so the terminology and the numerical value are inconsistent.
- [Figure 3 and accompanying text] The text says that without considering losses the PD3 signal will be half of the PD2 level, but Eqs. (4) and (8) predict identical functional forms (2I(1+cos(...))) apart from the phase, so the origin of the 1/2 factor is not explained.
- [Conclusion and Table 1] The conclusion states a '40×' improvement, but Table 1 shows a 40× ratio only for the 2000 m coil; the ratios for 200 m and 1000 m are approximately 12× and 28×. The approximate statement '10×, 30×, and 40×' should be aligned with the table values.
Circularity Check
The reported 40x ARW improvement is a built-in consequence of assuming PD2 and PD3 carry identical noiseless signals; the simulation's 'noiseless' idealization makes the quantitative claim reduce to its inputs.
-
other
[Simulation section, paragraph describing PD2/PD3 temporal offset (after Table 1, before Table 2)]
"Photodetectors in Fig.- 2, PD2 and PD3 were used to calculate the rotation rate, with a 3-meter fiber length difference introduced to create a temporal offset between the signals. Otherwise, perfect temporal match makes the system noiseless, which will be addressed for the future papers in terms of detailed investigation of the NRPPS-IFOG system."
The paper attributes the ARW gain to 'the use of PD2 and PD3, which are actually same signal but receive π-phase-shifted quadrature components,' i.e., the two channels are assumed to have perfectly correlated signals. The quoted sentence then states the direct consequence: in the simulation, perfect temporal matching makes the system noiseless. Because no independent detector noise (shot noise, thermal noise, or decorrelated RIN) is modeled, the reported ARW values and the 10-40x improvement are generated by the ad hoc 3 m offset rather than predicted from a physical noise source. The quantitative central claim therefore reduces by construction to the input assumption that the two photodetectors see the same common-mode signal.
full rationale
The passive π/2 biasing derivation is self-contained: the Jones-matrix calculation (Eqs. 17-24) shows that a retarder with φr = π/4 produces a 2φr = π/2 phase shift after double pass, and simultaneous quadrature detection at PD2/PD3 follows algebraically from Eqs. (4) and (8). This is a design choice, not a fitted result, and no self-citation is load-bearing. The circularity concern is limited to the headline ARW claim. The paper explicitly says the simulation is noiseless when PD2 and PD3 are temporally matched and introduces a 3 m fiber delay to generate the plotted Allan-variance behavior. Since PD2 and PD3 are assumed to be 'actually same signal,' common-mode noise cancellation is built into the input; the 40x improvement is therefore an artifact of the simulation construction rather than an independent prediction. The passive-bias concept itself retains independent content, so the score is moderate (6), not extreme.
Assumptions & free parameters
free parameters (2)
- retarder phase phi_r =
pi/4 (eighth-wave plate)
- temporal offset fiber length =
3 m (also 20, 50, 100 m tested)
assumptions (3)
- domain assumption The non-reciprocal element rotates polarization by exactly +45 degrees for both beam directions and introduces no phase or loss.
- domain assumption The two photodetectors PD2 and PD3 measure perfectly correlated (common-mode) noise so that subtraction cancels all noise.
- domain assumption Optical components (PBS, retarder, mirror) are ideal and lossless.
Cite this review
Pith. "Pith review of Phase Biasing System for Optical Gyroscope Using Passive Non-Reciprocal Polarization Techniques." pith.science (2026). https://pith.science/paper/AMKUAZJG
@misc{pith2026250519331,
author = {Pith},
title = {Pith review of: Phase Biasing System for Optical Gyroscope Using Passive Non-Reciprocal Polarization Techniques},
year = {2026},
howpublished = {\url{https://pith.science/paper/AMKUAZJG}},
note = {Machine review of arXiv:2505.19331}
}
abstract
Interferometric Fiber Optic Gyroscopes (IFOGs) are widely used in precision navigation systems due to their high sensitivity, robustness, and solid-state nature. To ensure linear response and accurate angular velocity measurement, a fixed $\pi/2$ phase bias is typically introduced between the clockwise (CW) and counter-clockwise (CCW) beams using active modulation components. However, these active elements increase system complexity, power consumption, cost, and susceptibility to thermal drift and long-term degradation. In this work, we present a novel IFOG configuration that, to the best of our knowledge, achieves for the first time, simultaneous operation at two quadrature points ($\pi/2$ and $3\pi/2$), providing natural noise suppression without relying on active components. This is made possible through the integration of a Non-Reciprocal Polarization-Dependent Phase Shifter (NRPPS), which introduces a pure $\pi/2$ passive phase shift. We detail the optical architecture and provide theoretical modeling using Jones calculus to demonstrate how the NRPPS element introduces passive quadrature biasing. Simulation results show that the proposed NRPPS-IFOG achieves significantly improved sensitivity, with Angular Random Walk (ARW) values up to 40x lower than those of conventional IFOGs, depending on the fiber coil length. The design further leverages quadrature-phase signal detection and adjustable temporal offsets between photodetectors to enhance noise suppression. This passive biasing approach eliminates the need for active modulation, offering reduced power consumption, improved stability, and enhanced long-term reliability.
Figures
Figures from the paper (2 more)
Forward citations
Cited by 1 Pith paper
-
A Simple and Novel Passive Double-Sensitivity Optical Gyroscope Based on Non-Reciprocal Polarization Techniques
The DS-NRPPS-IFOG combines a passive polarization phase shifter with a double-pass fiber coil to operate at two quadrature points, yielding simulated angular random walk up to 50x lower than a conventional double-sens...
Reference graph
Works this paper leans on
-
[1]
Fiber optic gyros for space, marine, and aviation applications
Glen A Sanders, Bogdan Szafraniec, Ren-Young Liu, Clarence L Laskoskie, Lee K Strandjord, and George Weed. Fiber optic gyros for space, marine, and aviation applications. Fiber Optic Gyros: 20th Anniversary Conference, 2837:61–71, 1996
1996
-
[2]
The fiber-optic gyroscope: Challenges to become the ultimate rotation-sensing technology
Hervé C Lefèvre. The fiber-optic gyroscope: Challenges to become the ultimate rotation-sensing technology. Optical Fiber Technology, 19(6):828–832, 2013
2013
-
[3]
Fiber optic sensor technology: an overview
Kenneth TV Grattan and T Sun. Fiber optic sensor technology: an overview. Sensors and Actuators A: Physical, 82(1-3):40–61, 2000
2000
-
[4]
L’éther lumineux démontré par l’effet du vent relatif d’éther dans un interféromètre en rotation uniforme
Georges Sagnac. L’éther lumineux démontré par l’effet du vent relatif d’éther dans un interféromètre en rotation uniforme. CR Acad. Sci., 157:708–710, 1913
1913
-
[5]
The fiber-optic gyroscope
Herve C Lefevre. The fiber-optic gyroscope. Artech house, 2022
2022
-
[6]
Temperature dependence of the polarization, dielectric constant, damping constant and the relaxation time close to the ferroelectric-paraelectric phase transition in linbo3
A Kiraci and HASAN Yurtseven. Temperature dependence of the polarization, dielectric constant, damping constant and the relaxation time close to the ferroelectric-paraelectric phase transition in linbo3. Optik, 132:183– 191, 2017
2017
-
[7]
The compensation of y waveguide temperature drifts in fog with the thermal resistor
Ying Li Wang, Li Yong Ren, Jin Tao Xu, Jian Liang, Meng Hua Kang, Kai Li Ren, and Nian Bao Shi. The compensation of y waveguide temperature drifts in fog with the thermal resistor. Advanced Materials Research, 924:336–342, 2014
2014
-
[8]
Research on multiple-integrated- optic-chip of re-entrant fiber optic gyro
Xianhui Mao, Qian Tian, Enyao Zhang, Liqun Sun, Yunhe Teng, and Jinjiang Liu. Research on multiple-integrated- optic-chip of re-entrant fiber optic gyro. In Opto-Ireland 2005: Optical Sensing and Spectroscopy, volume 5826, pages 636–644. SPIE, 2005. 9 arXiv Template A PREPRINT
work page 2005
Show all 23 references
-
[9]
Influence of the modulation phase error on bias error in interferometric fiber optic gyroscope
Ran Bi, Xuan She, Xiaowu Shu, and Cheng Liu. Influence of the modulation phase error on bias error in interferometric fiber optic gyroscope. Optical Engineering, 57(8):087104–087104, 2018
2018
-
[10]
Study of modulation phase drift in an interferometric fiber optic gyroscope.Optical Engineering, 49(11):114401–114401, 2010
Wei Wang and Junlong Wang. Study of modulation phase drift in an interferometric fiber optic gyroscope.Optical Engineering, 49(11):114401–114401, 2010
2010
-
[11]
Study on the drift of modulated phase in interference fiber optic gyroscope
Feng Sun, Lihui Wang, Gang Wang, and Gang Liu. Study on the drift of modulated phase in interference fiber optic gyroscope. J. Comput., 5(3):394–400, 2010
2010
-
[12]
Simultaneous suppression of thermal phase noise and relative intensity noise in a fiber-optic gyroscope
Nobuyuki Takei, Martin Miranda, Yuki Miyazawa, and Mikio Kozuma. Simultaneous suppression of thermal phase noise and relative intensity noise in a fiber-optic gyroscope. IEEE Sensors Journal, 23(3):2249–2254, 2023
2023
-
[13]
Modeling of thermal phase noise in a solid core photonic crystal fiber-optic gyroscope
Ningfang Song, Kun Ma, Jing Jin, Fei Teng, and Wei Cai. Modeling of thermal phase noise in a solid core photonic crystal fiber-optic gyroscope. Sensors, 17(11):2456, 2017
2017
-
[14]
Passive fiber optic gyroscope
Gert F Trommer, Hans Poisel, Wolfhardt Buhler, Engelbert Hartl, and Reinhard Muller. Passive fiber optic gyroscope. Applied optics, 29(36):5360–5365, 1990
1990
-
[15]
Passive fog imu for short-range missile application: from qualification toward series production
Gert F Trommer, R Mueller, and S Opitz. Passive fog imu for short-range missile application: from qualification toward series production. In Fiber Optic Gyros: 20th Anniversary Conference, volume 2837, pages 124–131. SPIE, 1996
1996
-
[16]
Fiber-optic gyroscope with [3 × 3] directional coupler
Sang K Sheem. Fiber-optic gyroscope with [3 × 3] directional coupler. Applied Physics Letters, 37(10):869, 1980
1980
-
[17]
Single polarization optical fiber and its applications
H Kajioka and H Matsumura. Single polarization optical fiber and its applications. Hitachi review, 33(4):215–218, 1984
1984
-
[18]
Orthogonal polarization fiber gyroscope with increased stability and resolution
CR Doerr, K Tamura, M Shirasaki, HA Haus, and EP Ippen. Orthogonal polarization fiber gyroscope with increased stability and resolution. Applied optics, 33(34):8062–8068, 1994
1994
-
[19]
Passively biased fiber-optic gyroscope and current sensor, March 16 2010
Hung-chia Huang. Passively biased fiber-optic gyroscope and current sensor, March 16 2010. US Patent 7,679,753
2010
-
[20]
A kind of implementation method of pi/2 phase bias of optical fiber interferometer, 2009
Bo Jia. A kind of implementation method of pi/2 phase bias of optical fiber interferometer, 2009. CN-101608930- A
2009
-
[21]
Passive topologically biased sagnac interferometer as a rotational sensor capable of sensing magnitude and direction of rotation, November 21 2024
Farhad Hakimi and Hosain Hakimi. Passive topologically biased sagnac interferometer as a rotational sensor capable of sensing magnitude and direction of rotation, November 21 2024. US Patent App. 18/198,230
2024
-
[22]
Energy-efficient optic gyroscope devices, November 29 2016
Xiaotian Steve Yao. Energy-efficient optic gyroscope devices, November 29 2016. US Patent 9,506,759
2016
-
[23]
All pm fiber laser mode locked with a compact phase biased amplifier loop mirror
Tongxiao Jiang, Yifan Cui, Pei Lu, Chen Li, Aimin Wang, and Zhigang Zhang. All pm fiber laser mode locked with a compact phase biased amplifier loop mirror. IEEE Photonics Technology Letters, 28(16):1786–1789, 2016. 10
2016
Reviewed August 7, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.