JPH0754257B2 - Optical fiber gyro scale factor stabilization method - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to an optical fiber gyroscope that measures a rotation angle or a rotation angular velocity of a moving body by utilizing the Sagnac effect, and particularly to stabilize the scale factor thereof. The present invention relates to implementation of a suitable method.

[Conventional technology]

As is well known, an optical fiber gyro is an interferometer type laser gyro interference system utilizing the Sagnac effect, which is constructed by an optical fiber, and a light beam divided into two via a beam splitter from both ends of the optical fiber loop. By making each incident light incident on the optical fiber loop by the optical fiber loop again into the same optical path through the beam splitter, to obtain optical interference based on the Sagnac effect of the optical fiber loop. There is. This interference light is taken out as an electric signal indicating rotation information (angular velocity) in the loop direction of the optical fiber loop through an appropriate photo detector.

By the way, in such an optical fiber gyro, various ideas for improving the sensitivity have been proposed.
The phase of the light propagating through the optical fiber loop is modulated as required, and the rotation information of the optical fiber loop is extracted as information that is phased by π / 2, that is, as sin function information (usually, the signal indicating the rotation information is the optical fiber loop. , Which is output as cos function information indicating the maximum value in the stationary state), is attracting attention as a method that can improve the resolution and stabilize the zero point. FIG. 3 shows an example of an optical fiber gyro employing such a phase modulation method.

That is, in the optical fiber gyro shown in FIG. 3, 1 is a laser beam generator, 2a and 2b are optical couplers acting as the beam splitters, 3 is an optical fiber loop (optical fiber coil), and 4 is PZT (electrostrictive vibration). Child)
And the like, 5 is a photodiode used as the photodetector, and 6 is the phase modulator 4
An oscillator for generating an electric signal having a frequency f ₀ for driving the frequency f ₀ , and a frequency f ₀ outputted from the oscillator 6 for a signal photoelectrically converted by the photodiode 5.
Is a lock-in amplifier that locks in based on the signal
Of the interference light components photoelectrically converted by the photodiode 5, it is a modulation component of the fundamental frequency.

E ₀ = KJ ₁ (η) 2sin2Δθ (1) where, η = 2mfsin2πf ₀ T, where K: constant J ₁ : Type 1 Bessel cell function-next term 2Δθ: Sagnac phase difference mf: modulation index f ₀ : modulation frequency It is known that the component of T: light propagation time in the optical fiber is extracted as its output E ₀ . By the way, according to the Bessel cell table, it is known that the value of J ₁ (η) becomes the maximum at η = 1.84. Therefore, in such an optical fiber gyro, η = 1.84 is usually obtained. After the modulation index mf (generally determined by the voltage value applied to the phase modulator 4) is determined, the sine wave component shown in the above equation (1) is taken out.

As described above, according to the optical fiber gyro of the phase modulation system as illustrated in FIG. 3, the interference light output extracted as the cos function information as shown in FIG. 4 (a) when the phase modulation is not performed. By the above phase modulation,
It is possible to take out as sin function information as shown in FIG. 4 (b). As is clear from comparing FIGS. 4 (a) and 4 (b), this means that the linearity of the optical fiber gyroscope in the low speed rotation range is improved, and the measurement sensitivity in the low speed rotation range is also greatly improved. Means that.

[Problems to be Solved by the Invention]

As described above, the optical fiber gyro by the phase modulation method improves the linearity in the low speed rotation range,
Although the sensitivity in the same range can be greatly improved, the scale factor, that is, the output value (measured value) fluctuates greatly with respect to the same angular velocity, so that it remains a problem in practical use.

The causes of such scale factor fluctuations are (a) the polarization of the optical fiber is unstable.

(B) Expansion and contraction of the optical fiber due to changes in ambient temperature.

(C) Instability of the modulation depth of the phase modulation circuit caused by these (a) or (b).

Incidentally, the modulation depth is generally determined by the magnitude of the voltage applied to the phase modulator and its frequency (modulation frequency). However, the optical fiber or the optical fiber loop wound around the phase modulator is usually determined. If the optical fiber that composes the optical fiber is disturbed by (a) or (b), a constant modulation depth can be obtained even if the voltage or frequency applied to the phase modulator is held constant. Absent.

And so on.

The present invention has been made in view of these circumstances, and an object thereof is to provide a method for stabilizing a scale factor of an optical fiber gyro that can eliminate the above-mentioned fluctuation factors and obtain a stable scale factor.

[Means for Solving the Problems]

In the present invention, when attention is paid to a specific waveform component of the photoelectric conversion signal extracted through the photo detector,
It should be noted that even when the amplitude or the convergence value varies, the functional form is always maintained in a similar relationship. The converted signal is taken out to obtain its functional form, and then the phase corresponding to the value represented by the product of the modulation index and its coefficient for obtaining the desired phase modulation depth is obtained based on the obtained functional form. While obtaining the modulation voltage or the phase modulation frequency, each time, referring to the value of the previously obtained phase modulation voltage or the phase modulation frequency, so that the value of the product of the modulation index and its coefficient is kept constant,
The phase modulation voltage or phase modulation frequency is controlled.

[Action]

It is as described above that the functional form of the extracted DC term component is kept in a similar relationship regardless of the change of the amplitude or the convergence value thereof, but this makes the desired phase modulation depth If the phase modulation voltage or the phase modulation frequency corresponding to the value to be obtained, that is, the value represented by the product of the modulation index and the coefficient is obtained in advance based on the extracted and obtained function form, It means that, by controlling the obtained value to be constant, the phase modulation with the same depth will be performed in any case thereafter. Of course, if the phase modulation depth can be controlled to be the same for such one waveform component (for example, DC term component), the phase of other waveform components (for example, fundamental wave component) actually extracted as the angular velocity output will also be obtained. The modulation depth is kept the same. In addition, the fact that the phase modulation depth is kept the same in this way means that each unstable element mentioned as a factor of the scale factor variation described above is substantially absorbed and eliminated by this modulation control. Means

〔Example〕

FIG. 1 shows an example of a device to which an embodiment of a method for stabilizing a scale factor of an optical fiber gyroscope according to the present invention is applied. In FIG. 1, the same elements as those shown in FIG. 3 are designated by the same reference numerals.

First, in order to facilitate understanding of the embodiment,
Referring to FIG. 1, consideration will be given to the angular velocity output by the phase modulation type optical fiber gyro.

Now, when the radius of the optical fiber loop 3 is R, the total length of the optical fiber forming the loop 3 is L, and the angular velocity when the loop 3 rotates in the loop direction is Ω, the position of the interference light due to the Sagnac effect is assumed. The phase difference Δθ is Is well known. Where λ
Is the wavelength of light propagating through the optical fiber loop 3,
C is the speed of light of the same propagation light.

Further, when the optical signal e _L generated from the beam generator 1 additionally shown in FIG. 1 is expressed by e _L = K sin ωt (3) K: constant ωt: angular frequency of light, The optical signal ec, which is divided into two by the optical couplers 2a and 2b and is incident on the optical fiber loop 3 in the CW direction (clockwise direction), is ec = k ₁ sin ωt (4) k ₁ : the optical path constant Similarly, the optical signal ecc, which is divided into _{two and} is incident on the same optical fiber loop 3 in the CCW direction (counterclockwise direction), becomes ecc = k ₂ sin ωt (5) k ₂ : the optical path constant, and these optical signals ec And ecc are phase-modulated by the phase modulator 4 and undergo a sag-nack phase shift through the optical fiber loop 3, and the optical signals ecm and ecc are returned from the optical fiber loop 3 to the optical coupler 2b.
m is ecm = k ₁ sin (ωt + α + Δθ) (6) eccm = k ₂ sin (ωt-β-Δθ) (7) Here, α and β represent the modulation amounts given to the optical signals ec and ecc by the phase modulator 4, respectively, and α = mf sin (pt + γ) (8) β = mf sin (pt−γ) ) (9) mf: Modulation index (value that indicates the depth of modulation) pt: Angular frequency to be modulated r: Phase difference with respect to angular frequency and details.

Now, in such an optical fiber gyro (gyro part 10), the optical signals ecm and eccm represented by the above formulas (6) and (7) are guided to the same optical path through the optical couplers 2b and 2a, and are It is considered that the diode 5 is synthetically input (incident). Here, when this combined light (interference light) is P _L , the interference light P _L is given by the following equation.

On the other hand, from the above equations (8) and (9), α + β = mf [sin (pt + γ) + sin (pt + γ)] = 2mf sin pt cos γ (11) α-β = mf [sin (pt + γ) -sin (pt + γ) )] = −2mf cos pt sin γ (12) Therefore, by substituting these equations (11) and (12) into the above equation (10), P _L = k ₁ k ₂ [1-cos (2ωt + 2mf sin pt cos γ)] × [1 + cos (2Δθ−2mf cos pt sin γ)] = k ₁ k ₂ [1-cos 2ωt cos (2mf cos r sin pt) −sin 2ωt sin (2mf cos γ sin pt) ] [1 + cos 2 Δθ cos (2 mf sin γ cos pt) + sin 2 Δθ sin (2 mf sin γ cos pt)] (13) The interference light P _L thus given is converted into a current signal through the photodiode 5 (photo-electric conversion). To be done. At this time, since the photodiode 5 does not respond to the angular frequency ωt of light, if this converted current is I _L , the term that does not include cos ωt and sin ωt in the above equation (13), that is, I _L = k ₁ k ₂ [1 + cos 2 Δθ cos (2mf sin γ cos pt) + sin 2 Δθ sin (2mf sin γ cos pt)] (14). Expanding this equation (14) into a series, Therefore, it can be seen that the current I _L is configured by synthesizing each harmonic component. For the phase modulation type optical fiber gyro, the fundamental wave component in this equation (15), that is, “2 sin 2 ΔθJ ₁ (2mf sin γ) cos pt” is usually used.
The term is extracted and used as its angular velocity output ((1) above)
See formula).

Let us now consider the value of sin r in Eq. (15).

In the above “2 sin 2 Δθ J ₁ (2mf sin γ) cos pt”, in order to keep the extracted output to the maximum, “J ₁ (2mf si
n γ) ”needs to be maximized. The best way to do this is to maximize sin γ and minimize the value of mf to be 2mf sin γ = 1.84. At this time, J ₁ (1.84) becomes the maximum.
On the contrary, decrease sin γ and increase mf.
Setting J ₁ (1.84) means increasing the voltage applied to the phase modulator (PZT) 4, which causes large unwanted mode vibration of the modulator and also increases modulation noise.

Thus, if sin γ is not maximized, that is, γ =
Unless it is (π / 2), the result of S / N is not good.

Therefore, in practice, this is also applied to the above equations (8) and (9). And need to. This means that the modulation phase moves symmetrically within the time when the light beam propagates in the optical fiber loop 3. Finally, let us consider the optimum value of the modulation frequency f ₀ based on these conditions.

Now, assuming that the optical propagation time T in the optical fiber loop 3 is determined as T = nL / C (16) n: core refractive index L: total length of optical fiber C: speed of light, the modulation phase becomes symmetrical. The condition for movement is T / 2N = nL / C (17) N = 1,2,3, ....

On the other hand, if the modulation frequency is f, this is f = 1 / T (18).

Therefore, if the above equation (17) is substituted into this equation (18), then f = C / 2NnL (19) and the minimum modulation frequency f ₀ that can satisfy the above γ = (π / 2) is , F ₀ = C / 2nL (20).

Next, the principle of this embodiment will be described with reference to FIG.

Now, pay attention to the DC term in the above equation (15). This output is
Since it is a component of the output of the photodiode 5 that does not have a modulation frequency component in the carrier, it can be easily extracted by using a low-pass filter. Then, the value is expressed by I = k ₁ k ₂ [1 + cos ₂ ΔθJ ₀ (2mf sin γ)] (21) as is clear from the equation (15). When the sagnat phase difference Δθ is zero (that is, when cos2Δθ = 1), its functional form is as shown in FIG. 2 (a).

If the value of the same DC term during the operation of the optical fiber gyro is expressed with cos2Δθ as a parameter, the amplitude changes as shown in FIG. 2 (b). That is, in this case, each functional form has I =
It shows a shape that converges while wrapping around k ₁ k _2.
The values of points q ₁ , q ₂ , q ₃ … on (2mf sin γ) that intersect with ＝ k ₁ k ₂ do not change.

On the other hand, usually it shall apply light Fuaibajiyairo, change in the laser output, or variations in the losses and extinction ratio of the optical system, change a branching ratio of the optical coupler, the so occurs, k ₁ above (15)
The value of k ₂ fluctuates. Therefore, in this case, the value (function form) of the DC term also has a convergent value as shown in FIG. 2 (c), but for example, with respect to k ₁ k ₂ = 1 and k ₁ k ₂ = 0.75, The values of the points q ₁ , q ₂ , q _3, ... On (2mf sin γ) where these functions intersect k ₁ k ₂ = 1 or k ₁ k ₂ = 0.75 remain unchanged in this case as well.

In this way, focusing on only a specific component of the photoelectric conversion output, the functional forms are similar even when the amplitude or the convergence value changes. Therefore, for example, one functional form as a sample for a specific component (DC term component) as shown in FIG.
Based on this functional form, for example, the AC applied voltage (2mf variable element) or modulation frequency (sin γ variable element) to the phase modulator (PZT) 4, which can be 2mf sin γ = 1.84, is obtained in advance, and then the above sample is used. Whenever the function form is calculated, the applied voltage or the modulation frequency is converted and controlled so that the relationship with 2mf sin γ = 1.84 is maintained. Phase modulation with the same depth is realized. Of course, if such phase modulation with the same depth is realized, the unstable element described above is absorbed thereby to stabilize the scale factor, and as a result, it is extracted as an angular velocity output. The reproducibility of the signal of the arbitrary component is also significantly improved.

Now, the device shown in FIG. 1 is configured as a device that preferably realizes one embodiment of the scale factor stabilizing method based on such a principle, and as shown in FIG.
The gyro unit 10 considered previously, and the modulation control unit 20 for driving the phase modulator (PZT) 4 of the gyro unit 10,
The signal processing unit 30 for obtaining an angular velocity output or an angular output by processing the electric signal photoelectrically converted through the photo diode 5 (usually a Pin photo diode) of the gyro unit 10 as required, The scale factor controller 40, which extracts the DC term component described in the above principle from the electro-converted signal and executes the required control based on the above principle for stabilizing the scale factor, is roughly divided into four parts. Is configured.

Among them, the modulation control unit 20, the (20) oscillator 21 the f ₀ oscillates and outputs the optimum modulation n times the frequency of the signal of frequency f ₀ as shown in formula, f ₀ to the oscillator signal The frequency of the signal (amplitude value) of the signal that has been received and finally is divided by 1 / n, and the frequency f ₀ is obtained by this frequency division.
, A programmable amplifier 23 that variably amplifies based on a control command CD applied to its control end, and an output amplifier 24 that further amplifies the variably amplified signal to a phase modulator 4 of the gyro unit 10, respectively. It is composed of As described above, the modulation control section 20 can freely control the modulation voltage value applied to the phase modulator 4 through the programmable amplifier 23. The signal divided by the frequency dividing circuit 22 is
It is also given to the signal processing unit 30, and is also used as a synchronization signal when the desired signal component is extracted (coherent detection) in the signal processing unit 30.

Further, the signal processing unit 30 includes an I / V (current / voltage) converter 31, which converts a current I _L (see the formula (15)) photoelectrically converted through the photodiode 5 of the gyro unit 10 into a voltage signal, And the frequency dividing circuit based on the converted voltage signal
Based on the frequency-divided signal (synchronization signal) by 22, for example, the amplification of the fundamental wave component (J ₁ component in the above equation (15), more specifically, “2 sin 2 Δθ J ₁ (2mf sin γ) cos pt”) The signal corresponding to the value k ₁ k ₂ 2sin ₂ ΔθJ ₁ (2mf sin γ)) is extracted (synchronous detection), and the angular velocity information of the gyro unit 10 or the angle information obtained by time integration of this is extracted based on this extracted signal. And a measurement signal processing circuit 32 for outputting By the way, like this example, light-
When the fundamental wave component of the electrical conversion signal is extracted and its measured value (angular velocity) is output, this characteristic is as shown in FIG. 4 (b).

Then, the scale factor control unit 40 outputs a signal corresponding to the DC term component “k ₁ k ₂ [1 + cos ₂ ΔθJ ₀ (2mf sin γ)]” of the equation (15) from the voltage conversion signal from the I / V converter 31. Low-pass filter 41 that extracts and filters only the components,
An amplifier 42 whose gain is set so that the filtered output is maintained at a predetermined value when not modulated by the phase modulator 4, and this amplifier output is converted into a digital signal with appropriate resolution.
A / D (analog / digital) converter 43, this A / D
An arithmetic circuit (CPU) 44 for arithmetically outputting the control command CD to be given to the programmable amplifier 23 in the mode described in detail below based on the converted output, and the above-mentioned arithmetic function by the arithmetic circuit 44. The sample voltage value and the applied voltage value to the phase modulator 4 (more accurately, as the angular velocity output) which can satisfy the modulation depth (for example, 2mf sin γ = 1.84) for obtaining a desired component appropriately. Programmable amplifier 23 for obtaining the voltage value
And a memory 45 for storing control command values).

The programmable amplifier 23 and the arithmetic circuit 44 will be described below.
The operation of the exemplary apparatus as a whole will be described with reference to FIG.

First, as a process for obtaining the phase modulation characteristic (functional sample) of the gyro unit 10 at that time, the programmable amplifier 23 is caused to change the output AC voltage for phase modulation from zero to a sufficiently large value. A simple scan command is given as the control command CD. This scanning command may be given through an appropriate external circuit, or may be given as a command preprogrammed in the arithmetic circuit 44 through the arithmetic circuit 44. At this time, the value of sin γ, which is the coefficient of the modulation index mf, is kept constant (for example, sin γ = 1). Such processing means scanning the value of 2mf in FIG. 2 (a) from zero to a sufficiently large value, whereby the arithmetic circuit 44
A signal showing its function form (the phase modulation characteristic of the gyro unit 10 at that time point) corresponding to the waveform shown in FIG. In the arithmetic circuit 44, the value of 2mf having the maximum value and the value of 2mf having the minimum value of the function form are respectively calculated based on the captured signal (function form), and the convergence value determined in the function form is obtained. 2mf corresponding to the intersection with the same function form
2mf corresponding to these maximum and minimum values, respectively.
Of the ratio up to the value of, and these are temporarily stored in the memory 45, and especially at the initial stage, from the value of 2mf corresponding to the maximum value and the minimum value, for example, 2mf sin γ
= 1.84 and the value of the voltage applied to the phase modulator 4 for obtaining the desired phase modulation depth (more precisely, the control command value to the programmable amplifier 23 for obtaining the voltage value), This is also temporarily stored in the memory 45.

In this way, when the control value (control command value for the programmable amplifier 23) based on the phase modulation characteristic of the gyro unit 10 at that time is determined, the phase modulator 4 is driven based on the determined control value. Based on this, the angular velocity measurement process through the gyro unit 10 and the signal processing unit 30 is executed. That is, as a result, some angular velocity output at that time is obtained.

When one angular velocity output is obtained in this way, a scan command is given to the programmable amplifier 23 again, and the same phase modulation characteristic extraction and control value determination operation based on the extracted characteristic are repeated. However, when determining the control value for the second time and thereafter, the arithmetic circuit 44 monitors the value of the obtained ratio, and if this is maintained,
That is, when it is determined that the function form extracted this time is similar to the previous function form, the control value stored in the memory 45 is maintained at the previous value and the same ratio value is obtained. This control value is updated only when there is a difference.

When the control value is updated in this way, the angular velocity measurement process to be executed next is also performed based on the phase modulation voltage (2 mf) corresponding to this updated content. Therefore, if the control value (phase modulation voltage 2 mf) is updated through the arithmetic circuit 44 in the direction in which the similar condition of the function type sampled each time is satisfied, it is output in this measurement process. The values of the angular velocities are always reproducible based on the phase modulation conditions of the same depth. Of course, even at this time, the predetermined condition of, for example, 2 mf sin γ = 1.84 is satisfied. This is no different from the fact that the updating of the control value (phase modulation voltage 2 mf) absorbs variations such as expansion and contraction of the optical fiber wound around the phase modulator 4.

As described above, in this device, the processing up to the control value determination including the scan of the phase modulation voltage 2mf and the angular velocity measurement processing based on the determined control value are alternately performed, thereby performing The reproducibility of the measured value (angular velocity output) is improved. This is based on the assumption that the processing will be performed by a computer.
It is possible to process in a cycle of S, that is, 10 mS.

In the above example, of (2mf sin γ), the value of sin γ is kept constant, and the value of (2mf sin γ) is kept constant by positively controlling it only for the value of 2mf. Although the case of keeping is described, as described in the previous principle, to keep the value of (2mf sin γ) constant, sin
Of course, the value of γ may be positively controlled.
However, in practice, this sin γ can take only a value from 0 to 1, so in order to obtain the required control range, the voltage applied to the phase modulator (PZT) 4, that is, the value of 2mf, is sufficiently large. You need to do it. As described above, the control of sin γ is the control of the phase modulation frequency f ₀ .

Further, in the above-mentioned embodiment, the light by the photodiode 5 is
Assuming that only the fundamental wave component of the electrical conversion output, that is, the J ₁ term is extracted, the value of (2mf sin γ) above is (1.
It has been explained that it is preferable to select 84) in order to bring the amplitude to the maximum, but other than that, for example, from the photoelectric conversion output, two components of J ₁ term and J ₂ term ((15) above) When extracting and using (expression), each (2mf sin
It is effective to choose the value of γ) near (2.4) in order to make the amplitude values equivalent and to minimize the DC term component (J ₀ term). Is clear from. Therefore, in such a case, the (2mf sin) selected to obtain the desired phase modulation depth is
The value of γ) is also preferably set near (2.4).
Extracting the two components J ₁ and J ₂ in this way is effective in expanding the measurement range of such an optical fiber gyroscope, and further, the tan and cot components are extracted from these components. The calculation is also effective in canceling and eliminating the noise component accompanying the expansion of the measurement range. However, the scale factor stabilizing method according to the present invention is not limited to the measurement signal processing circuit 32.
It goes without saying that the configuration is not limited.

〔The invention's effect〕

As described above, according to the scale factor stabilizing method for an optical fiber gyroscope according to the present invention, various unstable elements generated during phase modulation are well absorbed, and a uniform phase modulation depth can be obtained in any case. Like As a result, the reproducibility of the signal of the arbitrary component extracted as each speed output is significantly improved.

[Brief description of drawings]

FIG. 1 is a block diagram showing an example of a device for realizing an embodiment of a method for stabilizing a scale factor of an optical fiber gyroscope according to the present invention, and FIG. 2 is an explanation of functional similarity as a principle of the present invention. Fig. 3 is a block diagram showing an example of a conventional phase modulation type optical fiber gyro, and Fig. 4 compares the operational characteristics of the optical fiber gyro with and without phase modulation. FIG. 1 ... Beam generator, 2a, 2b ... Optical coupler, 3 ... Optical fiber loop, 4 ... Phase modulator, 5 ... Photodiode, 21 ... Oscillator, 22 ... Dividing circuit, 23 ... Programmable amplifier, 31 …… I / V converter, 32 …… Measurement signal processing circuit, 41 …… Low-pass filter, 42 …… Amplifier, 43
…… A / D converter, 44 …… Arithmetic circuit, 45 …… Memory.

Claims (3)

[Claims] 1. A light beam divided into two parts via a beam splitter is made to enter from both ends of an optical fiber loop, and the respective propagating lights of the incident light are phase-modulated while propagating the propagating lights. In the optical fiber gyro which again guides to the same optical path via the beam splitter and interfering light based on the Sagnac effect of the guided light is photo-electrically converted and extracted through the photo detector as the loop direction rotation information of the optical fiber loop, the phase modulation A first procedure for obtaining this functional form by extracting the specific waveform component from the photoelectric conversion output while appropriately scanning the phase modulation voltage or the phase modulation frequency that contributes to the determination of the modulation depth. Based on the obtained function form, the product of the modulation index and its coefficient to obtain the desired phase modulation depth. Of the phase modulation voltage or phase modulation frequency corresponding to the value of, and each time the first and second procedures are executed, the value of the previously obtained phase modulation voltage or phase modulation frequency. And a third procedure for controlling the phase modulation voltage or the phase modulation frequency so that the value of the product of the modulation index and its coefficient is kept constant. A method for stabilizing the scale factor of a fizzy gyro. 2. The method for stabilizing a scale factor of an optical fiber gyro according to claim 1, wherein the specific waveform component extracted from the photoelectric conversion output is a DC component. 3. The method according to claim 2, wherein the first to third procedures are executed in advance every time an angular velocity measurement value based on the photoelectric conversion output is calculated and output. Stabilization method for scale factor of optical fiber gyro.