JPH11257971A - Optical fiber gyro - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a digital optical tuber gyro in which effect of optical spike noise, occurring at the time of phase shift due to a modulation signal, can be eliminated. SOLUTION: A digital signal processor 100 synthesizes a signal for modulating the phase with frequency m/2τ (m is an integer larger than 1) and amplitude θ, assuming the propagation time of light through an optical fiber loop 6 is τ, a signal for switching the operating point of phase modulation at frequency 1/2τ and amplitude γ (=kπ, k is an integer of 1 or above), and a stepwise serrodyne signal having duration τ/n (n is an integer) at each stage such that phase shift +θ due to the phase modulation signal will be synchronous with phase shift +γ due to the phase modulation operating point switching signal.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an optical fiber gyro for detecting a rotational angular velocity or a rotational angle applied to, for example, an aircraft, a ship, an automobile, etc., and more particularly, to cancel a Sagnac phase shift by a stepped serrodyne signal. The present invention relates to a digital optical fiber gyro configured in a closed loop.

[0002]

2. Description of the Related Art In an optical fiber gyro, a change in light intensity of interference light obtained by recombining two lights propagating in an optical fiber loop in opposite directions is caused by a phase difference of Sagnac between the two lights. Then, cos (φs) is obtained. Therefore, as shown in FIG. 26, when the Sagnac phase difference is close to 0, the light intensity changes only slightly with respect to the change in the phase difference.

An optical fiber gyro that solves this problem is disclosed in Japanese Patent Application Laid-Open No. 56-94680.

In this technique, a phase modulator for phase-modulating two lights propagating through an optical fiber loop is provided, and as shown in FIG. 26, a phase shift of + π / 2 and -π / 2 is set to two.
By applying a phase modulation signal of a rectangular wave generated at a period of τ (τ: propagation time of light in an optical fiber loop) to the phase modulator, a change in the light intensity of the interference light is represented by cos (φs ± π /
It can be taken out in 2).

In a conventional digital fiber optic gyro, a serrodyne signal, which is a staircase signal having a duration of τ, is input to a phase modulator in order to expand the dynamic range of Sagnac phase difference detection. The phase modulation is performed so as to generate a phase difference having the same amount and a different sign as the Sagnac phase difference corresponding to the input angular velocity to the optical fiber gyro. Thus, a closed loop is configured to cancel the Sagnac phase difference between two lights propagating in the optical fiber loop. However, since the output of the serrodyne signal cannot be increased indefinitely, it is usually reset when the phase shift due to the serrodyne signal reaches ± 2π.

A digital optical fiber gyro is
A photodetector for converting interference light into a current signal, a current / voltage converter for converting the current signal to a voltage signal, a voltage amplifier for amplifying the voltage signal, and an A / D for converting the amplified voltage signal to a digital signal It comprises a converter, a digital signal processor for performing various signal processing using the digital signal, a D / A converter for outputting a phase modulation signal and a serrodyne signal, and the like.

[0007] When performing the A / D conversion, there is widely known a technique of performing sampling by limiting a signal band based on a sampling theorem. This band limitation is an extremely effective method for performing more stable closed-loop operation and rotational angular velocity detection in order to attenuate high-frequency electromagnetic noise superimposed on a signal.

In the prior art, when phase modulation is performed at a period of 2τ, a signal band is limited at a frequency of 1 / 2τ according to the sampling theorem, and a frequency twice as high as 1 / τ is applied.
Is configured to perform sampling.

If the serrodyne signal is not reset correctly, that is, if the modulation gain contains an error, a scale factor error occurs in the relationship between the input rate and the output rate of the optical fiber gyro.

[0010] As a digital optical fiber gyro for solving this problem, there are those described in Japanese Patent Application Laid-Open Nos. 3-210417 and 3-48715.

[0011] The former, as a phase-modulated signal to be input to the phase _{modulator, δφ 1 = φ 0 δφ 2} = aφ 0 δφ 3 = -φ 0 δφ 4 = -aφ 0 ( where, phi ₀ is a constant phase shift , A is cos φ ₀ = c
os (a positive constant that satisfies aφ ₀ ) is used as a signal having four steps as one cycle.

Here, the cycle of each step of the phase modulation signal is, for example, τ / 2 (2τ in total). A is 2. In this case, the modulation phase difference is as shown in FIG.

On the other hand, the latter uses a signal obtained by calculating a gain signal obtained from a phase difference electric signal and a total phase modulation signal as a phase modulation signal input to the phase modulator. This total phase modulation signal is obtained by summing the following signals.

A serrodyne signal created using the first frequency;

A velocity bias signal consisting of a periodic square wave having a first frequency.

A gain bias signal consisting of a series of step voltages transitioning at the end of successive equal intervals equal to half the period of the serrodyne signal.

Here, for example, the continuous equal interval time is τ / j (j is an integer). When j is 2, the speed bias signal has a frequency of 1 / 2τ. The velocity bias signal is a square wave that transitions between π / 2 and −π / 2, and the gain bias signal has a series of induced phase shifts of 2π, 0, −2π, and 0. In this case, the modulation phase difference is, for example, 0 to τ / 2 period: + π / 2 + 2π = + 5π / 2 τ / 2 to τ period: + π / 2 + 0 = + π / 2 τ to 3τ / 2 period: -π / 2- 2π = −5π / 2 3τ / 2−2τ period: −π / 2 + 0 = −π / 2, and therefore, as shown in FIG.

In both cases, the point of the light intensity at which the Sagnac phase difference becomes zero can be periodically shifted by applying a signal for shifting the phase at a predetermined period to the phase modulator. For this reason, the sensitivity of the light intensity near the Sagnac phase difference of 0 can be improved.

Further, as shown in FIGS. 27 and 28, the phase modulator applies ± iπ at a predetermined period (in FIG. 27, ± 4 when a = 2).
By applying a signal that causes a phase shift of (/ 3π, ± 2π in FIG. 28), a modulation gain error generated at the time of the phase shift can be detected.

Therefore, by performing modulation control such that the modulation gain error becomes zero and controlling the modulation gains of the phase modulation signal and the serrodyne signal to optimal values, the scale factor error in the gyro output generated by the serrodyne signal is reduced. Can be reduced.

[0021]

In a digital optical fiber gyro in which phase modulation is performed by a rectangular wave, a peak point (lightest point) or a bottom point (darkest point) of interference light is shifted for a short time during a phase shift. , A spike-like noise (light spike) is generated in the output waveform.

The optical spike noise has no relation to the Sagnac phase difference signal for performing the serrodyne control or the modulation gain error signal for performing the modulation control, and is an unnecessary signal.

In the digital optical fiber gyro described in JP-A-3-210417 and JP-A-3-48715, the direction (shape of the waveform) of the optical spike generated at the time of phase shift by a phase modulation signal is changed. (Dull part) is different. For this reason, there is a problem that an error due to a light spike included in the detection light signal cannot be efficiently removed.

Hereinafter, this problem will be described in detail.

In the technique described in Japanese Patent Application Laid-Open No. 3-210417, as described above, the phase modulation signal input to the phase modulator is δφ ₁ = φ ₀ δφ ₂ = aφ ₀ δφ ₃ = −φ ₀ δφ ₄ = A signal having one cycle of four steps of -aφ ₀ is used. Therefore, the light spikes contained in the interference light to be detected, as shown in FIG. _{27, φ 0 → aφ 0,} -φ 0 → -aφ 0
And light spike that occurs when the phase shift, aφ ₀ → -
The directions (top to bottom or bottom to top) at the end of the phase shift are different from each other depending on the optical spike generated when the phase shift is φ ₀ , −aφ ₀ → φ ₀ .

Japanese Patent Application Laid-Open No. Hei 3-210417 discloses that
It is described that a detection optical signal including these optical spikes is subjected to a frequency band limitation of 1 / 2τ according to the sampling theorem, and then sampled by an A / D converter.

FIG. 29 shows a signal waveform input to the A / D converter when the band is limited to 1 / 2τ by a primary low-pass filter.

This signal waveform is a case where both the Sagnac phase difference and the modulation gain error are zero, and an AC signal having a period of τ appears. This occurs as a result of the optical spike waveform being dulled by the band limiting filter.

Note that SP1 to SP4 in the figure are examples of sampling points by the A / D converter.

The modulation gain error is the difference between samples τ / 2 apart, that is, (SP1-SP2), (SP3-
Since it is detected by the demodulation processing of SP4), the influence of the optical spike cannot be removed. For this reason, modulation control is not performed accurately, and a modulation gain error occurs. As a result, there arises a problem that a gyro scale factor error cannot be sufficiently suppressed.

Even when the band is not limited, the waveform is changed due to the delay time and settling time of the D / A converter and the like, and the filter effect of the photodetector and the current / voltage converter for detecting the change in the light intensity of the interference light. Dulling can cause similar problems.

Also, in the technique described in Japanese Patent Application Laid-Open No. 3-48715, as shown in FIG. 28, 5π / 2 → π / 2,
The phase shift ends when an optical spike occurs at a phase shift of 5 / 2π → π / 2 and an optical spike occurs at a phase shift of π / 2 → −5π / 2, −π / 2 → 5π / 2. The directions at the time are different from each other.

Japanese Patent Laying-Open No. 3-48715 discloses a first band-pass filter for passing only a 1 / 2τ frequency band in order to detect a Sagnac phase difference in a detection optical signal containing these optical spikes. After performing band limitation by, sampling is performed by the first A / D converter,
The document states that, in order to detect a modulation gain error, the band is limited by a second band-pass filter that passes only the 1 / τ frequency band, and then sampling is performed by a second A / D converter.

FIG. 30 shows a signal waveform input to the A / D converter when the band limitation of 1 / τ is performed by the primary low-pass filter.

This signal waveform is a case where both the Sagnac phase difference and the modulation gain error are set to zero. An AC signal having a period of τ is generated as a result of the optical spike waveform being dulled by the band limiting filter. is there. JP-A-3-
Japanese Patent No. 48715 discloses a Sagnac phase difference detection unit and a modulation gain error detection unit that are separately provided and band-limited according to each signal. can be τ. However, as is clear from FIG. 30, the modulation gain error is detected by the difference between samples separated by τ / 2, that is, by the demodulation processing of (SP1−SP2) and (SP3−SP4). The effect cannot be sufficiently eliminated, and a modulation gain error occurs. As a result, there arises a problem that a gyro scale factor error cannot be sufficiently suppressed. Further, Japanese Unexamined Patent Publication No.
In JP-A-48715, since a Sagnac phase difference detecting section and a modulation gain detecting section for detecting and processing a signal at high speed are individually provided, there is a problem that the circuit becomes large, high in cost, and high in power. .

As described above, if the direction of the optical spike included in the interference light obtained by recombining two lights propagating in the optical fiber loop in opposite directions to each other at the end of the phase shift is not constant, A The optical spike becomes dull due to the band-limiting filter necessary for the / D conversion, and is superimposed on the A / D sampling signal, thereby causing an error in modulation control, and accurate various phase modulation signals and serrodyne signals are not generated. As a result, a scale factor error occurs in the gyro output.

Even when the band is not limited, the waveform is changed due to the delay time and settling time of the D / A converter and the like, and the filter effect of the photodetector and the current / voltage converter for detecting the light intensity change of the interference light. Dulling and similar errors occur.

According to the technique described in Japanese Patent Application Laid-Open No. 3-48715, a velocity bias signal transitioning between π / 2 and -π / 2 and a series of induced phase shifts of 2π, 0, -2π and 0 are provided. Depending on the combination with the gain bias signal having the following equation, the modulation phase difference is, for example, 0 to τ / 2 period: -π / 2 + 0 = -π / 2τ / 2 to τ period: -π / 2 + 2π = 3π / 2τ. ３3τ / 2 period: + π / 2 + 0 = + π / 2 3τ / 2 to 2τ period: + π / 2−2π = −3π / 2 In this case, the direction of the optical spike at the end of the phase shift becomes constant (all directions from bottom to top). However, as shown in FIG. 28, depending on the combination of the velocity bias signal and the gain bias signal, the direction of the optical spike at the end of the phase shift may not be constant. Errors cannot always be efficiently removed.

In a high-precision optical fiber gyro,
It is desired to reduce the random walk value. Here, the random walk value is a value indicating an S / N ratio of a light intensity signal of the interference light, and is defined by an execution value of a random noise component included in the signal.

The article "Intergrated Optic" by HC Lefevre et al., Introduced at the 1986 SPIE conference
s: A Practical Solution For Fiber-Optic Gyroscop
e "and US Pat. No. 5,530,545, the amplitude of the phase modulation signal for minimizing the random walk value is π /
It is reported to be in the range of 2 to less than π.

However, the random walk value is a function using the light intensity value of the interference light as a parameter.
If this light intensity value changes, the phase modulation amplitude for minimizing the random walk value also changes. In the digital optical fiber gyro described in the above-mentioned prior art, and in US Pat. No. 5,530,545, the output characteristics of a light source used to generate two lights propagating in opposite directions through an optical fiber loop, and the like, are changed. No consideration is given to a change in the light intensity value of the interference light due to the influence of the above, and the phase modulation signal is generated so as to have a predetermined phase modulation amplitude. For this reason, the random walk value cannot be sufficiently reduced.

The present invention has been made based on the above circumstances, and an object of the present invention is to eliminate the influence of an optical spike generated at the time of phase shift by a modulation signal, thereby achieving accurate modulation control. To provide an optical fiber gyro capable of improving the scale factor characteristic of the gyro output.

Another object of the present invention is to provide a digital optical fiber gyro that can always generate a phase modulation signal so that the phase modulation amplitude minimizes the influence of the random walk.

[0044]

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, the present invention is directed to a Sagnac phase difference of interference light obtained by recombining two lights propagating in opposite directions through an optical fiber loop. An optical fiber gyro for detecting a rotation angular velocity or a rotation angle, and a phase modulation signal,
A signal generating means for generating a composite signal of an operating point switching signal for phase modulation and a serrodyne signal for canceling a Sagnac phase difference; and a signal propagating through the optical fiber loop according to the composite signal generated by the signal generating means. And a phase modulator for phase-modulating each of the two lights to be performed, wherein the signal generation unit sets the propagation time of the light in the optical fiber loop to τ.
And a phase modulation signal of a rectangular wave for causing the phase modulator to perform phase modulation such that the phase difference of the interference light is: frequency: m / 2τ, m is an integer equal to or greater than 2; Frequency: 1 / 2τ Amplitude: γ = kπ, where k is an integer of 1 or more, a square wave phase modulation operating point switching signal for causing the phase modulator to perform phase modulation, and duration of each step: The combined signal of the stepped serrodyne signal and τ / n and n is an integer of 1 or more is synchronized so that the phase shift of + θ by the phase modulation signal and the phase shift of + γ by the operation point switching signal of the phase modulation are synchronized. Is generated.

In the present invention, a phase modulation signal satisfying the above conditions and a phase modulation operating point switching signal are used. Then, both are combined so that the phase shift of + θ (−θ) by the phase modulation signal and the phase shift of + γ (−γ) by the operation point switching signal of the phase modulation are synchronized.

Therefore, the phase shift end of the optical spike (spike noise generated by the phase shift) included in the interference light obtained by recombining the two lights propagating in the optical fiber loop in opposite directions to each other. The direction at the time (the portion where the waveform is dull) is always constant.

FIG. 1 shows an example of the modulation phase difference according to the present invention. Here, as the phase modulation signal, a signal that performs phase modulation so that the phase difference of the interference light becomes frequency: 1 / τ (m = 2) amplitude: 0 <θ ≦ π / 2 is used. The figure shows a modulation phase difference when a signal that performs phase modulation is used as the operating point switching signal so that the phase difference of the interference light is: frequency: 1 / 2τ amplitude: γ = π (k = 1) .

In the example shown in FIG. 1, γ + θ → γ−θ, γ−
θ → −γ−θ, −γ−θ → −γ + θ, −γ + θ → γ + θ
In the optical spike generated at the time of the phase shift, the direction at the end of the phase shift (the portion where the waveform is dull) is all the upward direction from the bottom.

In the detection optical signal including these optical spikes, after limiting the frequency band of ττ by the primary low-pass filter and then sampling by the A / D converter, 1 / FIGS. 31 and 32 show, as an example, a case in which sampling is performed by an A / D converter after the τ frequency band is limited.

These signal waveforms are obtained when both the Sagnac phase difference and the modulation gain error are set to zero.
An alternating signal that fluctuates every two appears. This is a result of the optical spike waveform being dulled by the band limiting filter.

Note that SP1 to SP4 in the drawing are examples of sampling points by the A / D converter. The Sagnac phase difference is the difference between samples τ apart,
Since it is detected by the demodulation processing of (SP1-SP3) and (SP2-SP4), the influence of the optical spike is canceled and eliminated through this processing.

The modulation gain error is the difference between samples separated by τ / 2, that is, (SP1-SP2), (SP3-S
Although the phase is detected by the demodulation process of P4), as shown in FIG. 1, since the phase modulation is applied so that the directions of the optical spikes are always in the same direction, the optical spikes after passing through the band limiting filter are applied. The effect of the dulling is almost always constant near the sampling point. This characteristic makes it possible to cancel out and remove the influence of the optical spike through the demodulation processing, perform more accurate modulation control, and reduce the gyro scale factor error.

The effect of the dulling of the optical spike is also substantially constant at points other than the sampling points shown in FIGS. 31 and 32, so that the sampling points can be arbitrarily selected. It has the advantage of being.

Further, even when oversampling a wideband signal without band limitation, even when a D / A converter or the like has a delay time and a settling time, a photodetector for detecting a change in the light intensity of interference light, a current receiver, and the like. Although the waveform becomes dull due to the filter effect in the voltage converter, etc., the effect of the optical spike is almost symmetrical for each τ / m, so as described above, the effect of the optical spike must be canceled out and removed. It is possible to perform accurate modulation control and reduce a gyro scale factor error. Further, there is an advantage that the sampling point can be arbitrarily selected. As described above, by performing the phase modulation so that the direction of the optical spike is always constant, the error due to the optical spike included in the detection optical signal becomes constant, and the error is canceled and removed through the demodulation process. Modulation control can be performed, and a gyro scale factor error can be reduced.

In the present invention, the signal generating means may generate a phase modulation signal so as to change the phase modulation amplitude θ according to the light intensity value of the interference light.

For example, there is provided means for storing a table indicating the correspondence between the light intensity and the amplitude θ of the phase modulation, and by referring to the table, the amplitude θ of the phase modulation and the amplitude θ corresponding to the light intensity value of the interference light are provided. Thus, a phase modulation signal may be generated.

As described above, the random walk value is
This is a function using the light intensity value of the interference light as a parameter. If this light intensity value changes, the phase modulation amplitude for minimizing the random walk value also changes. Therefore, by performing the above, it is possible to always generate the phase modulation signal such that the phase modulation amplitude is such that the influence of the random walk is minimized.

In the present invention, the signal generation means may further include a serrodyne signal generation means, and reset means for resetting the generation of the serrodyne signal in the serrodyne signal generation means. In this case, when the accumulation result of the modulation phase difference due to the serrodyne signal reaches the first threshold, the reset means reduces the phase difference of the interference light to -2.
resetting the serrodyne signal generation means so that the phase shift becomes π, and when the accumulation result of the modulation phase difference due to the serrodyne signal reaches a second threshold value lower by 2π than the first threshold value, the interference light The serrodyne signal generating means may be reset so that the phase difference becomes + 2π.

In this case, for example, by setting the first threshold value to + π and the second threshold value to -π, a signal based on a cellodyne signal is used as in the digital type optical fiber gyro described in the related art. Compared to the case of resetting to 0 when the accumulation result of the modulation phase difference reaches ± 2π,
The peak-to-peak value of the serrodyne signal can be reduced. Thus, power consumption can be reduced.

Further, for example, the first threshold is set to +2
By setting π and the second threshold value to 0, a single power supply can be used as a power supply for generating a serrodyne signal.

[0061]

DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below.

FIG. 4 is a schematic configuration diagram of a digital optical fiber gyro to which one embodiment of the present invention is applied.

Here, reference numeral 500 denotes an optical interferometer. The optical interferometer 500 includes a light source 1, a coupler 2, a polarizer 3, a coupler 4, a phase modulator 5, and an optical fiber loop 6 formed by winding an optical fiber a plurality of times.

The light source 1 includes a superluminescent laser diode (SLD) having a long coherence length, an SLD
An erbium-doped optical fiber light source (EDFS) having a higher output intensity than that is used.

Generally, the polarizer 3, the coupler 4, and the phase modulator 5 are composed of an optical integrated circuit (IOC).
ptical circuits) are integrated on one substrate. Also, in FIG. 4, the phase modulator 5 is provided at one end of the optical fiber loop 6, but it may be provided at both ends to perform phase modulation on the optical fiber loop 6 in opposite directions.

The light emitted from the light source 1 enters the coupler 4 via the coupler 1 and the polarizer 3, where it is split into two lights. One of these two lights propagates clockwise in the optical fiber loop 6, is phase-modulated by the phase modulator 5, and returns to the coupler 4. The other propagates counterclockwise through the optical fiber loop 6, returns to the coupler 4 after being phase-modulated by the phase modulator 5. Then, the two are multiplexed by the coupler 4. Thereby, interference light is formed.

Here, it is assumed that the rotational angular velocity Ω is applied to the optical fiber loop 6. At this time, an optical path length difference occurs between the two lights propagating in the optical fiber loop 6 in opposite directions, and as a result, a phase difference occurs. This phase difference is equal to the Sagnac phase difference φs
It is called. This Sagnac phase difference φs is expressed by the following equation.

Φs = (2πDL / λc) Ω (1) where D is the loop diameter of the optical fiber loop, L is the optical fiber length, λ is the wavelength of light emitted from the light source 1, and
c is the speed of light.

The intensity P of the interference light and the Sagnac phase difference φs have the following relationship.

[0070] _{P = (P 0/2)} (1 + cosφs) (2) where, P ₀ is the peak value of the intensity of the interference light.

The interference light formed by the coupler 4 is received by the light receiver 7 and converted into a current signal according to the intensity of the interference light. This current signal is converted into a voltage signal by a current / voltage (I / V) converter 8 and then converted to a wideband amplifier 9.
Amplified by

The voltage signal amplified by the amplifier 9 is converted by an A / D converter 32 for detecting light intensity into a crystal oscillator, FP
According to a sampling signal F from a reference signal generator 300 composed of a GA or the like, the signal is sampled and converted into a digital signal. It is considered that a change in the light intensity value of the interference light due to a change in the output of the light source 1 or the like is due to a temperature or aging. Therefore, if sampling is on the order of several msec, it is possible to detect a change in the light intensity value of the interference light. For this reason, A for light intensity detection
As the / D converter 32, a low-speed, low-power-consumption, small-sized and low-cost A / D converter can be used.

The voltage signal amplified by the amplifier 9 is
After a DC component is removed by a DC remover 20 constituted by a high-pass filter and a DC offset adder, the DC component is amplified by a wideband amplifier 22. Thereafter, the signal is input to an A / D converter 26 for serrodyne control and modulation control, where it is sampled according to a sampling signal D from a reference signal generator 300 and converted into a digital signal.

The sampling method is based on the sampling theorem, which limits the band of the signal to 1 / (2T) [Hz] and then performs sampling at a frequency of 1 / T [Hz]. There is an oversampling method in which a wide band is used without any particular limitation and sampling is performed at a frequency sufficiently higher than the band.

The latter involves the problems that the number of samples and the signal processing are greatly increased, and that the latter is susceptible to high-frequency electromagnetic noise.

On the other hand, in the former, since the number of samples and signal processing can be reduced, and high-frequency electromagnetic noise can be easily removed by a band limiting filter, a more stable rotation angular velocity or rotation angle can be obtained. This is an extremely effective method in a digital optical fiber gyro for detection.

The signal band is limited by the current / voltage converter 8,
Any of the amplifier 9 and the amplifier 22 may be performed.

The optical fiber gyro uses various optical fiber loop diameters D and optical fiber lengths L according to the required performance. In general, from a low-precision optical fiber gyro to a high-precision optical fiber gyro, the optical fiber length L is about 100 m to 2 km. The propagation time τ of these optical fiber loops is about 0.5 μ
s to 10 μs.

Since the A / D converter 26 is required to have a sampling frequency of 1 / τ or more, even when the band is limited based on the sampling theorem, high-speed A / D conversion and signal processing are required. .

The digital signal processor 100 is composed of a computer such as a DSP or an integrated logic IC such as an ASIC or an FPGA. This digital signal processor 1
00 calculates the gyro output (angular velocity or angle) according to the Sagnac phase difference φs of the interference light.

When the propagation time of light in the optical fiber loop 6 is τ, the digital signal processor 100 determines that the phase difference of the interference light is: frequency: m / 2τ, and m is an integer of 2 or more Amplitude: θ And a rectangular wave phase modulation signal for switching the operating point of the phase modulation so that the frequency is 1 / 2τ, the amplitude is γ = kπ, and k is an integer of 1 or more. An operating point switching signal and a Sagnac phase difference φs corresponding to the input angular velocity to the optical fiber gyro 6 produce a phase difference of the same amount and a different sign, and the duration of each step is τ / n (where n is 1 or more) Integer), a stepped serrodyne signal,
Are combined with a phase shift signal of + θ (−θ) by the phase modulation signal and a + γ (−
γ) is generated in synchronization with the phase shift.

Here, the amplitude θ of the phase shift by the phase modulation signal is determined according to the light intensity value of the interference light so that the random walk value is minimized. As described in the problem to be solved by the invention, it is said that the amplitude of the phase modulation signal for minimizing the random walk value is in the range from π / 2 to less than π. The random walk value is a function using the light intensity value of the interference light as a parameter. If the light intensity value changes, the phase modulation amplitude for minimizing the random walk value also changes.

Therefore, in the present embodiment, the interferometer 500 is referenced with reference to a table prepared in advance showing the relationship between the light intensity value of the interference light and the phase modulation amplitude θ for minimizing the random walk value. The amplitude θ corresponding to the light intensity value of the interference light obtained from the detection result is determined.

When the length of the optical fiber of the optical fiber loop 6 is L, the refractive index of the optical fiber constituting the optical fiber loop 6 is n ₀ , and the speed of light is c, the propagation time τ is expressed by the following equation.

Τ = n ₀ L / c (3) Further, since the output of the serrodyne signal cannot be increased indefinitely, the accumulation result of the modulation phase difference due to the serrodyne signal is equal to the first threshold value (for example, + π). When it reaches
When the serrodyne signal is reset so that the phase difference of the interference light becomes a phase shift of −2π, and the accumulation result of the modulation phase difference due to the serrodyne signal reaches a second threshold value lower by 2π than the first threshold value, the interference occurs. The serrodyne signal is reset so that the phase difference of the light becomes a phase shift of + 2π.

The composite signal (total of the phase modulation signal, the phase modulation operating point switching signal and the serrodyne signal) output from the digital signal processor 100 is converted into an analog signal by the D / A converter 10, and The signal is input to the phase modulator 5 via the driver 11.

In response to this, the phase modulator 5 converts the two lights propagating in the optical fiber loop 6 in opposite directions into a combined signal of a phase modulation signal, a phase modulation operating point switching signal, and a serrodyne signal. Each phase modulation is performed accordingly.

In this embodiment, a phase modulation signal that satisfies the above-mentioned conditions and a phase modulation operating point switching signal are used. Then, a phase shift of + θ (−θ) by the phase modulation signal and + γ (−γ) by the operation point switching signal of the phase modulation.
Are combined such that the phase shift of the two is synchronized.
For this reason, the direction at the end of the phase shift of the optical spike (spike noise generated by the phase shift) included in the interference light obtained by recombining the two lights propagating in the optical fiber loop in opposite directions. (The portion where the waveform becomes blunt) is always constant.

FIG. 1 shows a rectangular modulation signal which is phase-modulated as a phase modulation signal so that the phase difference of the interference light is: frequency: 1 / τ (m = 2) amplitude: 0 <θ ≦ π / 2 Using signals,
Modulation in the case of using a rectangular wave signal for performing phase modulation so that the phase difference of the interference light becomes: frequency: 1 / 2τ amplitude: γ = π (k = 1) as the operating point switching signal of the phase modulation The phase difference is shown.

In this case, γ + θ → γ−θ, γ−θ → −γ
The optical spike generated at the time of the phase shift of −θ, −γ−θ → −γ + θ, −γ + θ → γ + θ is the optical spike at the end of the phase shift as shown in the waveform of the intensity-time characteristic of the detection light in FIG. (Portions where the waveform is dull) are all directed upward from the bottom.

FIG. 2 is a diagram showing a phase modulation signal of a rectangular wave for performing phase modulation such that the phase difference of the interference light is as follows: frequency: 3 / 2τ (m = 3) amplitude: 0 <θ ≦ π / 2 Using signals,
Modulation in the case of using a rectangular wave signal for performing phase modulation so that the phase difference of the interference light becomes: frequency: 1 / 2τ amplitude: γ = π (k = 1) as the operating point switching signal of the phase modulation The phase difference is shown.

In this case, γ + θ → γ−θ, γ−θ → −γ
The optical spike generated at the time of the phase shift of −θ, −γ−θ → −γ + θ, −γ + θ → γ + θ is the optical spike at the end of the phase shift as shown in the waveform of the intensity-time characteristic of the detection light in FIG. (The part where the waveform is rounded) is the direction from bottom to top.

FIG. 3 shows a phase modulation signal of a rectangular wave for performing phase modulation such that the phase difference of the interference light becomes frequency: 2 / τ (m = 4) amplitude: π / 2 ≦ θ <π Using signals,
Modulation in the case of using a rectangular wave signal for performing phase modulation so that the phase difference of the interference light becomes: frequency: 1 / 2τ amplitude: γ = 2π (k = 2) as the operating point switching signal of the phase modulation The phase difference is shown.

In this case, γ + θ → γ−θ, γ−θ → −γ
The optical spike generated at the time of the phase shift of −θ, −γ−θ → −γ + θ, −γ + θ → γ + θ is the optical spike at the end of the phase shift as shown in the waveform of the intensity-time characteristic of the detection light in FIG. (The portion where the waveform is dull) is the direction from top to bottom.

FIGS. 1 to 3 are examples, but according to the present embodiment, the directions of the optical spikes (portions where the waveform is dull) at the end of the phase shift are all the same.

For example, although not shown, a rectangle for performing phase modulation as a phase modulation signal such that the phase difference of the interference light is: frequency: 1 / τ (m = 2) Amplitude: π / 2 ≦ θ <π Using wave signals,
When a rectangular wave signal for performing phase modulation is used as the operating point switching signal of the phase modulation so that the phase difference of the interference light is: frequency: 1 / 2τ amplitude: γ = 2π (k = 2), γ + θ → γ-θ, γ-θ → -γ-θ, -γ-θ →
As for the optical spikes generated when the phase shift is −γ + θ, −γ + θ → γ + θ, the directions of the optical spikes (portions where the waveform is dull) at the end of the phase shift are all directed downward from the top.

As described above, in the present embodiment, the direction of the optical spike (the portion where the waveform becomes dull) at the end of the phase shift is always constant. For this reason, the error due to the optical spike included in the detection light signal is always constant, and the error is canceled or removed by the demodulation processing in the modulation control, and more accurate modulation control can be performed. As a result, the scale factor error of the gyro can be reduced.

In the present embodiment, the light intensity value of the interference light is detected, and the phase modulation signal is generated so as to perform the phase modulation amplitude θ according to the detected value with reference to the table. By doing so, it is possible to always suppress the influence of the random walk.

Next, the digital signal processor 100 which is a main configuration of the present embodiment will be described in detail.

The components of the digital signal processor described below may be executed in hardware by an integrated logic IC such as ASIC or FPGA, or may be executed in software by a computer such as DSP. It may be something. Also, the specific processing in each configuration is
The present invention is not limited to those described below, but may be any as long as the same function can be realized.

First, the digital signal processor 100a in the case where m for specifying the phase modulation frequency (m / 2τ) by the phase modulation signal is an even number of 2 or more (for example, in FIGS. 1 and 3) will be described.

FIG. 5 is a schematic configuration diagram of the digital signal processor 100a when m for specifying the phase modulation frequency (m / 2τ) by the phase modulation signal is an even number of 2 or more.

As shown in FIG. 5, the digital signal processor 100a includes a signal processor 110a, a serrodyne controller 140, a modulation controller 170a, and a light intensity calculator 2a.
10 and a gyro output operation unit 240.

The signal processing section 110a is provided with the A / D converter 32
Adjust the gain of the signal output from. A / D
The gain of the signal output from the converter 26 is adjusted and demodulated.

FIG. 6 is a schematic configuration diagram of the signal processing section 110a shown in FIG.

The multiplier 136 outputs the output signal of the A / D converter 32 and the AGC (Auto
matic Gain Control). Thereby, the gain of the output signal of the A / D converter 32 is adjusted to be always constant, and the light intensity calculation unit 210 can perform stable light intensity calculation.

The multiplier 134 multiplies the output signal of the A / D converter 26 by the AGC signal from the light intensity calculator 210. Thus, the gain of the output signal of the A / D converter 26 is adjusted to be always constant, and the serrodyne control unit 140 and the modulation control unit 170a can perform stable signal generation.

The first demodulator 112a is connected to the reference signal generator 3
00, the frequency m / 2τ (m
Demodulates the output signal of the multiplier 134 with the reference signal B having the same frequency as that of the even signal of 2 or more. For example, in the example shown in FIG. 1, since m = 2, the output signal of the multiplier 134 is demodulated by the reference signal B having the frequency 1 / τ.

If the sampling interval in the A / D converter 26 is shorter than τ / m, that is, if there are two or more sampling points during the τ / m period, the A / D converter 2
6 and a filter for performing an averaging process are provided between the first demodulator 112a and the first demodulator 112a.
Output at intervals.

This demodulator is composed of, for example, a multiplier.

The second demodulator 114a is connected to the reference signal generator 3
The output signal of the first demodulator 112a is demodulated by the reference signal A generated at 00 and having the same frequency as the frequency 1 / 2τ of the operating point switching signal of the phase modulation signal. This demodulator is composed of, for example, a multiplier.

The τ delay unit 116a is connected to the second demodulator 114a
Is delayed by time τ.

An adder 118a adds the output of the second demodulator 114a and the output of the τ delay 116a. The result of the addition is a deviation signal of the serrodyne control system which is a closed loop including the serrodyne control unit 140. This deviation signal is
Generated in synchronization with the phase modulation signal (frequency m / 2τ), that is, at time intervals of τ / m,
Output to 40.

The τ delay unit 116a and the adder 118a
Constitutes a sequential demodulator.

Here, in the phase modulation operation shown in FIG. 1, it is assumed that an angular velocity is input to the optical fiber loop 6 and a Sagnac phase difference φs occurs. in this case,
The modulation phase difference is as shown in FIG.

As shown in FIG. 7, the A / D converter 2
6 is demodulated by the first demodulator 112a at the frequency of the phase-modulated signal, and the demodulation result and the demodulation result of the first demodulator 112a delayed by τ time by the τ delay unit 116a are simply added to obtain a cellodyne signal. Necessary for generation,
A signal (deviation signal of the serrodyne control system) corresponding to the Sagnac phase difference φs cannot be extracted.

Therefore, in the present embodiment, a second demodulator 114a that performs demodulation at the frequency of the operating point switching signal of the phase modulation is provided, and the frequency and phase of the phase modulated signal are output from the A / D converter 26. The demodulation is performed at the frequency of the modulation operation point switching signal, and the demodulation result and the demodulation result before τ time are added. In this way, as shown in FIG. 7, a signal corresponding to the Sagnac phase difference φs is extracted.

The signal processing unit 110a has a means for extracting a signal corresponding to the Sagnac phase difference φs.
A sequential demodulator consisting of a τ delay unit 116a and an adder 118a is used. Instead of the sequential demodulator, the output of the A / D converter 26 is used as the frequency of the phase modulation signal and the frequency of the phase modulation operating point switching signal. A demodulator composed of an arithmetic unit for averaging the result of demodulation at the time interval of 2τ may be used. Also in this case, as is clear from FIG. 7, a signal corresponding to the Sagnac phase difference φs can be extracted.

The operation unit 122a includes a reference signal generator 300
The averaging process is performed on the output signal of the first demodulator 112a in units of time specified by the averaging time signal C generated in. This calculation result is a deviation signal of a modulation control system that is a closed loop including the modulation control unit 170a. This deviation signal is output to modulation control section 170a.

Here, in the phase modulation operation as shown in FIG. 1, it is assumed that the phase modulation by the phase modulator 5 includes an error in the modulation gain G. In this case, the modulation phase difference is as shown in FIG.

In this embodiment, the output of the A / D converter 26 is demodulated by the first demodulator 112a at the frequency of the phase modulation signal, and the demodulation result is averaged by the arithmetic unit 122a at predetermined time intervals. Thus, a signal (deviation signal of the modulation control system) corresponding to the modulation gain error is extracted.

The deviation signal is a phase modulation signal generated by the modulation control section 170a when the modulation gain of the phase modulator 5 or the D / A converter 10 changes due to temperature change or aging. This is for adjusting the gain of the composite signal of the modulation operating point switching signal and the serrodyne signal. Therefore, the signal processing unit 110a and the modulation control unit 170
a and the servo loop including the interferometer 500 need not be fast. In addition, since the response of the arithmetic unit 122a to the modulation control unit 170a may be several Hz to several hundred Hz, the averaging time signal C may be on the order of msec. However, it is preferably an integer multiple of the frequency m / 2τ of the phase modulation signal.

As is clear from these descriptions, when m specifying the phase modulation frequency (m / 2τ) is an even number of 2 or more, the Sagnac phase difference is (m / 2τ) * (1/2)
τ), and are equivalently represented by sideband spectra of (m−1) / 2τ and (m + 1) / 2τ. On the other hand, the modulation gain error appears in the frequency band of m / 2τ.

For example, when m = 4, the Sagnac phase difference appears in the sideband spectral bands at 3 / 2τ and 5 / 2τ, and the modulation gain error appears in the 2 / τ frequency band.

The band is limited by the sampling theorem.
When the Sagnac phase difference and the modulation gain error are detected by one A / D converter, some band restrictions may be considered.

A method of performing band limitation at a frequency of (m-1) / 2τ and performing sampling at a frequency of (m-1) / τ or more, a method of performing band limitation at a frequency of m / 2τ and performing a band limitation at a frequency of m / 2τ or more Method of sampling at frequency, (m + 1) / 2τ
, And a method of performing sampling at a frequency of (m + 1) / τ or more.

When m = 2, the A / D converter input signal waveforms when the band is limited at a frequency of 1 / 2τ and when the band is limited at a frequency of 1 / τ are shown in FIGS. 31 and 32.
Shown in

These signal waveforms are obtained when both the Sagnac phase difference and the modulation gain error are set to zero.
An alternating signal that fluctuates every two appears. This is a result of the optical spike waveform being dulled by the band limiting filter.

[0129] SP1 to SP4 in the figure are examples of sampling points by the A / D converter. The Sagnac phase difference is the difference between samples τ apart,
Since it is detected by the demodulation processing of (SP1-SP3) and (SP2-SP4), the influence of the optical spike is canceled and eliminated through this processing.

The modulation gain error is the difference between samples τ / 2 apart, ie, (SP1−SP2), (SP3−SP2).
Is detected by the demodulation processing SP4), when placing the modulator gain error and .delta.G, signal corresponding to the modulation gain error after the demodulation process (error signal) is, _{approximately, (P 0/2) *} 2π * δ
It is represented by G.

The scale factor error of the gyro is several tens p
In order to reduce the value to pm or less, modulation gain control of 0.1% or less is required. To achieve this control accuracy, the deviation signal, it is necessary 0.0063 * (P _0/2) following the detection accuracy.

As shown in FIG. 1, since the phase modulation is applied so that the directions of the optical spikes are always in the same direction, the influence of the dulling of the optical spikes after passing through the band-limiting filter is affected by the vicinity of the sampling point. , And the sampled optical spike component is canceled and removed through demodulation processing.

The deviation signal obtained near each sampling point in FIG. 31 is obtained when the modulation gain error is zero.
In _{0.00235 * (P 0/2)} , in FIG. 32,
0.00097 is a * (P _0/2). These are 0.037% as errors in modulation gain control, respectively.
This corresponds to 0.015%, indicating that modulation gain control of 0.1% or less is possible. As a result, it is possible to reduce the gyro scale factor error.

Further, the influence of the dulling of the optical spike is also substantially constant at points other than the sampling points shown in FIGS. 31 and 32, so that the sampling point can be arbitrarily selected. Has the advantage of becoming

Further, even when the band limitation is not performed, the influence of the optical spike is substantially symmetrical for each τ / m, so that the influence of the optical spike can be canceled and removed as described above. It is.

One example is shown in FIG.

This is because in an optical fiber gyro having an optical fiber length L = 100 m and an optical fiber loop propagation delay time τ = 0.5 μs, the phase modulation frequency 1 / m = 2 is obtained.
τ, phase modulation operating point switching frequency 1 / 2τ, signal bandwidth 10 MHz, D / A converter settling time about 100
7 shows an A / D converter input signal waveform when ns is set.
Band limitation by sampling theorem is 1 / τ = 2MH
z or 1 / 2τ = 1 MHz, but in this waveform diagram, the signal band is 10 MHz, which is 5 to 10 times higher than that. Nevertheless, a very large AC signal that fluctuates every τ / 2 appears. This is a dull optical spike caused by the fact that the influence of the settling time of the D / A converter and the like cannot be ignored because the optical fiber length becomes short, and the dull optical spike even when the signal band is not limited. This becomes an error source of modulation gain control. Even in such a case, since the phase modulation is applied so that the directions of the optical spikes are always in the same direction,
As shown in FIG. 9, the effect of the dullness of the optical spike is almost constant near the sampling point, and the sampled optical spike component is canceled and removed through the demodulation processing. As a result, it is possible to reduce the gyro scale factor error.

Further, the influence of the dulling of the light spikes is also observed at points other than the sampling points shown in FIG.
Since the sampling point is wide and substantially constant, there is an advantage that the sampling point can be arbitrarily selected.

As described above, by performing the phase modulation so that the direction of the optical spike is always constant, the error due to the optical spike included in the detected optical signal becomes constant, and the above error is canceled and removed through the demodulation processing. Modulation control is performed, and a gyro scale factor error can be reduced.

Returning to FIG. 5, the description will be continued.

The serodyne control unit 140 is a signal processing unit 1
A serrodyne signal is generated according to a deviation signal corresponding to the Sagnac phase difference φs output from the adder 118a of 10a at a time interval of τ / m.

FIG. 10 shows the cellodyne control unit 1 shown in FIG.
It is a schematic block diagram of 40.

The first computing unit 142 includes a signal processing unit 110a
The deviation signal of the serrodyne control system sent from
At the time interval of. The result is a signal proportional to the input angular velocity to the optical fiber loop 6.

The second calculator 144 functions as an amplifier or a low-pass filter. The setting of the gain or the filter constant is performed according to the servo loop design of the serrodyne control system.

The third computing unit 146 integrates the output signal of the first computing unit 142 sent via the second computing unit 144. As described above, the output signal of the first computing unit 142 is a signal proportional to the input angular velocity to the optical fiber loop 6. If the input angular velocity is constant, the first computing unit 142
Is also constant. In this case, the third computing unit 146
Results in a staircase signal having a constant height of τ / m and a constant height. This staircase signal is output to the modulation control section 170a as a serrodyne signal.

It should be noted that the signal processor 110 a
In the case where a demodulator constituted by an arithmetic unit averaging at a time interval of τ is used, the deviation signal of the serrodyne control system sent from the signal processing unit 110a is sent to the first arithmetic unit 142 at a time interval of 2τ. The third computing unit 146 integrates the output signal of the first computing unit 142 sent via the second computing unit 144 at time intervals of τ. In this case, the output result of the third computing unit 146 is a staircase signal having a duration of τ of each staircase.

The comparator 150 determines whether the accumulation result of the modulation phase difference due to the serrodyne signal output from the third computing unit 146 has reached the first threshold value or the second threshold value 2π lower than the first threshold value. Determine whether or not.

When the first threshold value is reached, the third calculator 146 is reset so that the phase difference of the interference light becomes a phase shift of -2π. When the second threshold value is reached, the third calculator 146 is reset so that the phase difference of the interference light becomes a phase shift of + 2π.

The reference value storage section 148 stores values for specifying the first and second threshold values used in the comparator 150 and addition / subtraction values for resetting in the third modulator 146.

Here, the operation of the comparator 150 will be described in more detail with reference to FIGS.

FIG. 11 shows an operation flow of the comparator 150 when the first threshold value is set to + 2π and the second threshold value is set to −2π.

The comparator 150 first has a third computing unit 146
It is determined whether or not the accumulation result of the modulation phase difference due to the serrodyne signal generated in step 2 has reached 2π (step 100).
1).

If it reaches 2π, it is determined that a phase shift of 2π (negative reset) is to be performed (step 100).
2). Thereafter, a command is issued to the third computing unit 146 so that the phase difference of the interference light becomes a phase shift of -2π (step 1003). In response, the third computing unit 146
Adjusts the output of the serrodyne signal so that the phase difference of the interference light becomes a phase shift of −2π.

On the other hand, if it has not reached 2π, it is determined whether or not the accumulation result of the modulation phase difference by the serrodyne signal has reached -2π (step 1004).

If it reaches -2π, it is determined that a phase shift of + 2π (positive reset) is to be performed (step 10).
05). Thereafter, a command is issued to the third computing unit 146 so that the phase difference of the interference light becomes a phase shift of + 2π (step 1006). In response, the third computing unit 14
6 adjusts the output of the serrodyne signal so that the phase difference of the interference light becomes a phase shift of + 2π.

On the other hand, if it has not reached -2π, it is determined that resetting is unnecessary (step 1007). In this case, the phase shift command is not output to the third computing unit 146.

FIG. 12 shows an operation flow of the comparator 150 when the first threshold is set to + 2π and the second threshold is set to 0. This flow is different from the flow shown in FIG. 11 in that step 1004a is provided instead of step 1004, and it is determined whether or not the accumulation result of the modulation phase difference due to the serrodyne signal has become smaller than zero. If it is smaller than 0, step 1005
Otherwise, to step 1007.

FIG. 13 shows an operation flow of the comparator 150 when the first threshold value is set to + π and the second threshold value is set to -π. This flow differs from the flow shown in FIG. 11 in that steps 1001a and 1004b are provided instead of steps 1001 and 1004, respectively.

In step 1001a, it is determined whether or not the accumulation result of the modulation phase difference by the serrodyne signal has reached + π. If the value has reached + π, the flow shifts to step 1002; otherwise, the flow shifts to step 1004b.

In step 1004b, it is determined whether or not the accumulation result of the modulation phase difference by the serrodyne signal has reached -π. If the value has reached -π, the process proceeds to step 1005; otherwise, the process proceeds to step 1007.

FIG. 14 shows the waveform of the serrodyne signal output from the third computing unit 146. Here, FIG. 14A shows the waveform of the serrodyne signal generated by the third calculator 146 according to the flow shown in FIG. 11, and FIG. 14B shows the waveform of the third calculator 146 according to the flow shown in FIG. 3 shows a waveform of a serrodyne signal generated by the above. FIG.
(C) shows the waveform of the serrodyne signal generated by the third computing unit 146 according to the flow shown in FIG.

As can be seen from these waveforms, according to the flow of FIG. 12, a serrodyne signal can be generated with a single polarity regardless of the polarity of the input angular velocity to the optical fiber loop 6. For this reason, D /
The A converter and the driver can be configured with a single power supply.

Generally, the digital signal processor 100 and the reference signal generator 300 operate on a single power supply such as +5 V. Therefore, the fact that the D / A converter 10 can operate on a single power supply means that a common power supply is required. This has the advantage that the size and cost of the device can be reduced.

In the flow of FIG. 12, it is determined in step 1001 whether or not the accumulation result of the modulation phase difference by the serrodyne signal has become larger than 0.
In 04a, it may be determined whether or not the accumulation result of the modulation phase difference due to the serrodyne signal is -2π or less.
A similar effect is achieved.

According to the flow of FIG. 13, the peak-to-peak value of the serrodyne signal is reduced to half of that in the flow shown in FIG. This makes it possible to reduce power consumption at the terminating resistor of the phase modulator 5 to 1/4.

Returning to FIG. 5, the description will be continued.

The modulation control section 170a provides a rectangular wave phase modulation signal for performing phase modulation such that the light intensity of the interference light is frequency: m / 2τ, m is an even number equal to or greater than 2 and θ: Is generated by the serrodyne control unit 140 and a square wave phase modulation operating point switching signal for performing phase modulation such that the light intensity of the frequency is: ττ amplitude: γ = kπ, k is an integer of 1 or more The signal is combined with the cellodyne signal to generate a composite signal.

FIG. 15 shows a modulation control section 170a shown in FIG.
FIG.

The reference value storage unit 196 stores a reference output value necessary for causing a phase shift of θ (0 <θ ≦ π / 2 in the example shown in FIG. 1) and γ (in the example shown in FIG. 1, and a reference output value necessary for performing the phase shift of π).

The amplitude θ generator 202 includes a reference value storage unit 19
The reference output value necessary for causing the phase shift of θ stored in 6 to be performed is read out, and is corrected based on the optimum phase modulation amplitude value sent from the light intensity calculation unit 210. Then, a constant output signal having the corrected value is generated.

The amplitude γ generator 208 includes the reference value storage unit 19
A reference output value required for performing the phase shift of γ stored in 6 is read, and a constant output signal having this value is generated.

The phase modulation generator 204a receives the reference signal B (frequency: m / 2τ,
m is an even number of 2 or more), and modulates the constant output signal generated by the amplitude θ generator 202.

As a result, a rectangular wave phase modulation signal for performing phase modulation is generated so that the phase difference of the interference light has a frequency of m / 2τ, and m has an even amplitude of 2 or more.

The phase modulation generator 204a is formed of, for example, a multiplier.

The adder 206 includes a phase modulation generator 204a.
The constant output signal generated by the amplitude γ generator 208 is added to the phase modulation signal generated by the above.

The modulation generator 200a includes the reference signal generator 3
The output signal of the adder 206 is modulated by the reference signal A (frequency: 1 / 2τ) generated in 00. In particular,
The rising edge of the output signal of the adder 206 and the rising edge of the reference signal A are synchronized so that they are multiplied.

Accordingly, the phase difference between the rectangular wave phase modulation signal for performing phase modulation so that the phase difference of the interference light is frequency: m / 2τ, where m is an even number equal to or greater than 2, and the phase difference between the interference light Generates a composite signal of a rectangular wave phase modulation operating point switching signal for performing phase modulation such that frequency: 1 / 2τ amplitude: γ.

The adder 176a adds the phase modulation signal generated by the modulation generator 200a and the operating point switching signal of the phase modulation and the serrodyne signal generated by the serrodyne control section 140, and adds the phase modulated signal and the phase modulated signal. And a composite signal of the operating point switching signal and the serrodyne signal is generated.

The first computing unit 172 includes a signal processing unit 110a
This is an integrator that integrates the deviation signal of the modulation control system sent from the control unit.

The second computing unit 174 is an amplifier or a low-pass filter. The gain or the filter constant is designed in accordance with the servo loop design of the modulation control system.

The multiplier 178 outputs the signal output from the adder.
The gain of the combined signal of the phase modulation signal, the phase modulation operating point switching signal, and the serrodyne signal is adjusted according to the deviation signal of the modulation control system output from the second computing unit 174.
This result is output to the D / A converter 10.

The adder 198 includes an output signal of the amplitude θ generator 202 (a value corresponding to the amplitude θ of the phase modulation by the phase modulation signal) and an output signal of the amplitude γ generator 208 (the operating signal for switching the operating point of the phase modulation). (A value corresponding to the amplitude γ of the phase modulation). This addition result is output to the light intensity calculation unit 210.

FIG. 16 shows a modulation control section 170 shown in FIG.
3A shows the waveform of an output signal at each unit when m specifying the frequency of the phase modulation signal is 2. 16A shows the output signal waveform of the amplitude θ generator 202, FIG. 16B shows the output signal waveform of the phase modulation generator 204a, and FIG. 16C shows the output signal of the amplitude γ generator 208. 16D shows the output signal waveform of the adder 206, and FIG. 16E shows the output signal waveform of the modulator 200a.

Returning to FIG. 5, the description will be continued.

The light intensity calculator 210 is provided with a signal processor 110
a, the output signal of the A / D converter 32 for detecting the light intensity sent from the controller a, the value corresponding to the amplitude θ of the phase modulation by the phase modulation signal sent from the modulation controller 170a, and the operation of the phase modulation. The peak value P of the light intensity of the interference light is based on the sum of the value corresponding to the amplitude γ of the phase modulation by the point switching signal and the value.
_{Find 0} .

Then, by referring to a previously prepared table showing the correspondence between the peak value of the light intensity of the interference light and the optimum phase modulation amplitude for minimizing the random walk value, the table is determined according to the peak value P ₀ obtained. Determined amplitude value.

Further, an AGC signal is generated based on the peak value P ₀ of the light intensity of the interference light.

Further, the light intensity calculator 210 calculates the correspondence between the peak value of the light intensity of the interference light and the scale factor correction coefficient for correcting the input / output scale factor error of the optical fiber gyro. , A scale factor correction coefficient corresponding to the obtained peak value P ₀ is determined.

FIG. 17 is a block diagram showing the light intensity calculator 210 shown in FIG.
FIG.

The first computing unit 212 includes a signal processing unit 110a
A signal corresponding to the light intensity of the interference light detected by the interferometer 500 sent from the multiplier 136 of the
a based on the value corresponding to the amplitude θ of the phase modulation by the phase modulation signal and the value corresponding to the amplitude γ of the phase modulation by the operating point switching signal of the phase modulation sent from the adder 198 of a. , An arithmetic process for obtaining the peak value P ₀ of the light intensity of the interference light is performed.

This calculation process is performed by the following equation.

P ₀ = 2V _out / {1 + cos (γ ± θ)} (4) Here, V _out is a multiplier 136 of the signal processing unit 110a.
Is an output value of a signal corresponding to the light intensity of the interference light detected by the interferometer 500, which is sent from the controller. γ + θ or γ-θ
Is obtained from the output value of the signal sent from the adder 198 of the modulation control unit 170a.

It is not necessary to perform this operation at high speed. The processing may be performed at time intervals of about msec. Therefore, the signal from the multiplier 136 of the signal processing unit 110a is averaged at the above-mentioned time intervals, and the peak value P ₀ of the light intensity of the interference light is calculated by the above equation (4) using the averaged value. You may ask for it.

Although not shown, the second computing unit 214
The table stores the correspondence between the peak value of the light intensity of the interference light and the optimal phase modulation amplitude that minimizes the random walk value. Then, the optimum phase modulation amplitude value corresponding to the peak value P ₀ of the light intensity of the interference light calculated by the first calculator 212 is searched from the table. Then, it outputs the searched amplitude value to amplitude θ generator 202 of modulation control section 170a. In response to this, the amplitude θ generator 202 outputs the reference output for the phase modulation signal stored in the reference value storage unit 196 so that the phase modulation signal can perform phase modulation with the determined amplitude value. Correct the value.

The register 216 stores, as an initial value, the peak value of the light intensity of the interference light first calculated by the first computing unit 212 when the interferometer 500 is assembled.

The third calculator 218 obtains a difference between the peak value of the light intensity of the interference light obtained by the first calculator 212 and the initial value stored in the register 216.

The fourth computing unit 220 integrates the result of the third computing unit 218. The result of this integration is used as an AGC signal.
Output to the signal processing unit 110.

The fifth computing unit 222 is not shown, but
The table stores the correspondence between the peak value of the light intensity of the interference light and the scale factor correction coefficient for correcting the input / output scale factor error of the optical fiber gyro. Then, a scale factor correction coefficient corresponding to the peak value P ₀ of the light intensity of the interference light calculated by the first calculator 212 is searched from the table, and is output to the gyro output calculator 240.

Here, the relationship between the peak value of the light intensity of the interference light and the phase modulation optimum amplitude value that minimizes the random walk value will be described.

The relationship between the peak value of the light intensity of the interference light and the scale factor correction coefficient for correcting the input / output scale factor error of the optical fiber gyro will be described later.

A signal corresponding to the light intensity of the interference light detected by the optical fiber gyro includes a signal as described below.
Several types of random noise are superimposed.

First, the relative intensity noise of the light source 1 (RIN: Relative Int)
ensity noise) is superimposed on the output light of the interferometer 500.

[0203] RIN is defined as follows.

RIN [dB / Hz] ≡10 · log (δP ² / Pa ² / B) (5) N ′ _RIN [1 / (Hz ^1/2 )] ≡δP / Pa / B ^1/2 (6) Here, δP ² is the mean square value of the fluctuation of the light intensity,
Pa is the average light intensity, and B is the equivalent bandwidth defined only in the real number region.

The average intensity when an arbitrary phase modulation amplitude Ψ [rad] is applied is expressed by the following equation.

[0206] _{P a = (P 0/2} ) · (1 + cosΨ) (7) Accordingly, the random noise amount contained in the interferometer 500 of the output _{light, N'RIN · P a [Wrms} / (Hz 1/2 _{)] = N'RIN · (P} 0/2) · (1 + cosΨ) becomes (8).

Next, the output light of the interferometer 500 on which the RIN is superimposed is converted into a current signal by the light receiver 7, and at this time, the output light is correlated with the output value of the converted signal. Shot noise is superimposed.

The shot noise is defined by the following equation.

_{N ′ SN} [Arms / (Hz ^1/2 )] ≡ (2 · e · i) ^1/2 = (e · P ₀ · S · (1 + cosΨ)) ^1/2 (9) where e Is the electron charge (= 1.6 · 10)
^-19 [c]), i is the average current value [A] generated by converting light, and S is the sensitivity [A /
W]. _{N ′ SN} is the effective value of noise defined only in the real number domain.

Noise in Current / Voltage Converter 8 Next, the current signal on which shot noise is superimposed is converted into a voltage signal in the current / voltage converter 8. At this time, thermal noise of the used resistor and other current / voltage noise are superimposed.

The thermal noise and the current / voltage noise are expressed by the following equations.

N ′ _TN [Vrms / (Hz ^1/2 )] ≡ (4 · k · RT · ^1/2 ) (10) _N ′ _VN [Vrms / (Hz ^1/2 )] ≡ (V _N ² + (I _N · R) ² ) ^1/2 (11) where k is Boltzmann's constant (= 1.38 · 10
^-23 [J / ゜ K]), R is the resistance value [Ω] of the current / voltage converter 8, T is the absolute temperature [゜ K], and N is
' _TN is the effective value of noise defined only in the real number domain. V _N is the voltage noise of the operational amplifier, and I _N is the current noise. N ′ _VN is the effective value of noise defined only in the real number domain.

The voltage signal on which the above-described random noise is superimposed is sampled by the A / D converter 26, and thereafter, is subjected to serrodyne control, angular velocity (angle) calculation, and the like in the digital signal processor 100a. . The amount of random noise in the voltage signal is (8) to
From equation _{(11), V noise [Vrms / (Hz} 1/2)] = ((N'RIN · (P 0/2) · S · R (1 + cosΨ)) 2 + R 2 · e · P 0 · S · (1 + cosΨ) + N ′ _TN ² + N ′ _VN ² ) ^1/2 (12)

From this equation, a theoretical equation of a random walk value for an arbitrary phase modulation amplitude θ can be derived.

The optical fiber loop 6 has an angular velocity Ω [rad /
s] is input, whereby the Sagnac phase difference φs [r
ad] has occurred. As described above, in the present embodiment, the intensity of the interference light corresponding to the phase difference φs is transmitted through the optical fiber loop 6 in the opposite directions to each other.
The two lights are detected by phase-modulating them with an amplitude Ψ [rad].

If the random noise described above is not taken into account, the amplitude Ψ
P (+ Ψ), P (−Ψ) of the interference light phase-modulated by
Is represented by the following equation.

[0217] P (+ Ψ) = (P 0/2) · (1 + cos (φs + Ψ)) (13) P (-Ψ) = (P 0/2) · (1 + cos (φs-Ψ)) (14) Thus, demodulated signal V _out by synchronous _{detection, V out = (P (+} Ψ) -P (-Ψ)) · S · R · 1/2 = (P 0/2) · (cos (φs + Ψ) -cos (φs- Ψ)) · S · R · 1/2 = - (P 0/2) · S · R · sin (φs) · sin (Ψ) (15) where substitutes (15) to (1) _{Then, V out = - (P 0} /2) · S · R · sin {2π · D · L · Ω / (λ · c)} · sin (Ψ) (16) calculating the input angular velocity Omega from this equation Ω = λ · c / (2π · D · L) · sin ⁻¹ [V _out / ｛(− P _0/2 ) · S · R · sin (Ψ)}] (17)

The random walk value RWK of the gyro is
Since it is generally defined by the equivalent bandwidth ± 2B defined in the complex domain, if the real domain definition is converted to the complex domain definition by multiplying by 1 / (2 ^1/2 ), equation (12) is obtained. ,
From equation (17), it is expressed as follows.

[0219] ^{RWK [rad / (s 1/2)} ] = λ · c / (2π · D · L) · sin -1 [V noise / {(- P 0/2) · S · R · sin (Ψ )}] · {1 / ( 2 1/2)} = λ · c / (2π · D · L · 2 1/2) · sin -1 [{N'RIN · (P 0/2) · S · R · (1 + cos (Ψ ))} 2 + R 2 · e · P 0 · S · (1 + cos (Ψ)) + N'TN 2 + N'VN 2] 1/2 / {(- P 0/2) · S · R · sin (Ψ)｝ (18) RWK [゜ / (h ^1/2 )] = λ · c / (2π · D · L · 2 ^1/2 ) · sin ⁻¹ [｛N ′ _RIN · (P _{0/2) · S · R} · (1 + cos (Ψ))} 2 + R 2 · e · P 0 · S · (1 + cos (Ψ)) + N'TN 2 + N'VN 2] 1/2 / {(- P _0/2 ) · S · R · sin (Ψ)｝ · (3600) ^1/2 · 180 / π (19) From the above equations (18) and (19), Cal manner, random walk value is a function of the amplitude of the peak value P ₀ and the phase modulation of the light intensity as a parameter of the interference light, if Kaware the peak value P ₀ of the light intensity of the interference light, the random walk value The amplitude of the phase modulation that minimizes is also changed.

In FIG. 18, the relative intensity noise RIN of the light source 1 is -115 [dB / Hz], and the peak value P of the light intensity of the interference light is shown.
Random walk value when ₀ is ₂ to 50 μW−
9 shows phase modulation amplitude value characteristics. Here, the horizontal axis represents the modulation depth in the cos curve of the interference light characteristic. When γ = π, it represents the value of γ−θ, and when γ = 2π, it represents the value of θ.

According to this figure, it can be seen that the amplitude value of the phase modulation for minimizing the random walk value changes depending on the peak value P ₀ of the light intensity of the interference light. For example, P ₀ = 5
The optimum amplitude value at ₀ μW is about 170 degrees, but P ₀ = 2
It is 130 degrees in μW.

Therefore, in the present embodiment, the second computing unit 21
4 stores in advance a table showing the correspondence between the peak value P ₀ of the light intensity of the interference light and the optimum phase modulation amplitude value that minimizes the random walk value, which is obtained from the expressions (18) and (19). The optimal phase modulation amplitude value corresponding to the peak value P ₀ of the light intensity of the interference light calculated by the first computing unit 212 is searched from the table.

Returning to FIG. 5, the description will be continued.

The gyro output calculation section 240 calculates the input angular velocity or rotation angle to the optical fiber loop 6 according to the signal corresponding to the Sagnac phase difference φs sent from the first calculator 142 of the serodyne control section 140. .

FIG. 19 is a schematic configuration diagram of the gyro output operation section 240 shown in FIG.

The first arithmetic unit 242 is provided with the serrodyne control unit 1
The signal according to the Sagnac phase difference φs sent from the first computing unit 142 of the forty, that is, the signal corresponding to the input angular velocity to the optical fiber loop 6 is time-integrated. This result is proportional to the rotation angle of the optical fiber loop 6. When calculating the average angular velocity, the integration result may be further divided by the integration time.

The register 244 stores the input and output scale factor values in the initial state of the optical fiber gyro.

The second computing unit 246 uses the scale factor value stored in the register 244 and the scale factor correction coefficient sent from the fifth computing unit 222 of the light intensity computing unit 210 to generate the first computing unit 242. Correct the output. Thereby, the input angular velocity or rotation angle to the optical fiber loop 6 is calculated. This second computing unit 246 is formed of, for example, a multiplier.

Here, the relationship between the peak value of the light intensity of the interference light and the scale factor correction coefficient for correcting the input / output scale factor error of the optical fiber gyro will be described.

The output value of the optical fiber gyro (specifically, the angular velocity or angle calculated by the gyro output calculation section 240) changes in correlation with the change of the peak value P ₀ of the light intensity of the interference light. This is considered to be due to the following reasons.

The angle increment in the gyro output calculator 240 with respect to the input angular velocity Ω to the optical fiber loop 6 is represented by θ _angle
In this case, the angle increment θ _angle when the phase shift of 2π occurs due to the reset of the serrodyne signal is expressed by the following equation.

Θ _angle = n ₀ · λ / D (20) This equation is equivalent to using the duration τ of each step of the serrodyne signal, and θ _angle = c · λ · τ / (D / L) (21)

The peak value P ₀ of the light intensity of the interference light
Occurs when the loss in the optical path from the light source 1 itself or from the light source 1 to the optical receiver 7 via the optical fiber loop 6 changes, and their loss characteristics differ depending on the wavelength of light. In the present embodiment, as described above, the light source 1 used for the optical fiber gyro has a short coherence length, and the wavelength spectral characteristics are widely distributed over several tens to several hundreds of nm. Therefore, the light intensity changes due to the change in loss, and the effective center-of-gravity wavelength value of the light changes. As a result, the value of the center-of-gravity wavelength λ in Expressions (20) and (21) changes, and the angle increment θ _angle changes. This is considered to be an input / output scale factor error.

The inventors confirmed that the sensitivity of the scale factor was about several tens to several hundreds ppm for a change of the light intensity value of 10%. The light intensity value easily changes due to temperature, aging, and the like. This cannot be used for a high-performance optical fiber gyro requiring a scale factor error of several to several tens of ppm or less.

Thus, in the present embodiment, as described above, the fifth computing unit 222 of the light intensity computing unit 210
A table indicating the correspondence between the peak value of the light intensity of the interference light and the scale factor correction coefficient (correction coefficient of the scale factor value stored in the register 244) is stored in advance;
The scale factor correction coefficient corresponding to the peak value of the light intensity of the interference light calculated by the first calculator 212 of the light intensity calculator 210 is searched from the table. Then, in the second computing unit 246 of the gyro output computing unit 240, the scale factor value stored in the register is corrected using the searched correction coefficient (specifically, the scale factor value stored in the register is corrected to the scale factor value stored in the register). The result of the multiplication by the correction coefficient is multiplied by the result according to the rotation angle or the average angular velocity output from the first calculator 242 of the gyro output calculator 240).

In this manner, a scale factor error caused by a change in light intensity value is corrected.

The digital signal processor 100a in the case where m for specifying the phase modulation frequency (m / 2τ) by the phase modulation signal is an even number of 2 or more (eg, in FIGS. 1 and 3) has been described.

Next, in the case where m for specifying the phase modulation frequency (m / 2τ) by the phase modulation signal is an odd number of 2 or more (for example, in the case of FIG. 2), the digital signal processor 1
00b will be described.

FIG. 20 is a schematic configuration diagram of the digital signal processor 100b in a case where m for specifying the phase modulation frequency (m / 2τ) by the phase modulation signal is an odd number of 2 or more.

As shown in FIG. 20, the digital signal processor 100b includes a signal processor 110b, a serrodyne controller 140, a modulation controller 170b, and a light intensity calculator 2b.
10 and a gyro output operation unit 240.

Serrodyne control unit 140, light intensity calculation unit 2
10 and the gyro output operation unit 240 are the same as those used in the digital signal processor 100a shown in FIG. 5, and thus detailed description thereof will be omitted.

The signal processing section 110b includes the A / D converter 32
Adjust the gain of the signal output from. A / D
The gain of the signal output from the converter 26 is adjusted and demodulated.

FIG. 21 is a block diagram showing the signal processing unit 110 shown in FIG.
It is a schematic block diagram of b. The signal processing unit 1 shown in FIG.
The same components as those in 10a are denoted by the same reference numerals.

The multiplier 136 outputs the output signal of the A / D converter 32 and the AGC (Auto
matic Gain Control). Thereby, the gain of the output signal of the A / D converter 32 is adjusted to be always constant, and the light intensity calculation unit 210 can perform stable light intensity calculation.

The multiplier 134 multiplies the output signal of the A / D converter 26 by the AGC signal from the light intensity calculator 210. Thus, the gain of the output signal of the A / D converter 26 is adjusted to be always constant, and the serrodyne control unit 140 and the modulation control unit 170b can perform stable signal generation.

The first demodulator 112b is connected to the reference signal generator 3
00, the frequency m / 2τ (m
Is an odd number of 2 or more) by the reference signal B ′ having the same frequency as
The output signal of the multiplier 134 is demodulated. For example, in the example shown in FIG. 2, since m = 3, the output signal of the multiplier 134 is demodulated by the reference signal B ′ having a frequency of 3 / 2τ.

When the sampling interval in the A / D converter 26 is shorter than τ / m, that is, when there are two or more sampling points during the τ / m period, the A / D converter 2
6 and a filter for performing an averaging process are provided between the first demodulator 112b and the first demodulator 112b.
Output at intervals.

This demodulator is composed of, for example, a multiplier.

The second demodulator 114b is connected to the reference signal generator 3
The output signal of the first demodulator 112b is demodulated by the reference signal A having the same frequency as the frequency 1 / 2τ of the operating point switching signal of the phase modulation signal generated in 00. This demodulator is composed of, for example, a multiplier.

The τ delay unit 116b is connected to the first demodulator 112b
Is delayed by time τ.

The adder 118b adds the output of the first demodulator 112b and the output of the τ delay 116b. The result of the addition is a deviation signal of the serrodyne control system which is a closed loop including the serrodyne control unit 140. This deviation signal is
Generated in synchronization with the phase modulation signal (frequency m / 2τ), that is, at time intervals of τ / m,
Output to 40.

The τ delay unit 116b and the adder 118b constitute a sequential demodulator.

Here, in the phase modulation operation as shown in FIG. 2, it is assumed that an angular velocity is input to the optical fiber loop 6 and a Sagnac phase difference φs occurs. in this case,
The modulation phase difference is as shown in FIG.

The signal processing unit 110b differs from the signal processing unit 110a shown in FIG. 6 in that, as shown in FIG.
The output of the / D converter 26 is demodulated by the first demodulator 112b at the frequency of the phase modulation signal, and the demodulation result and the τ delay 1
By simply adding the demodulation result of the first demodulator 112b delayed by τ by 16b, a signal (deviation signal of a serrodyne control system) corresponding to the Sagnac phase difference φs required for generating a serrodyne signal can be taken out. it can.

In the signal processing section 110b, as means for extracting a signal corresponding to the Sagnac phase difference φs,
Instead of using a sequential demodulator consisting of a τ delay unit 116b and an adder 118b, the result of demodulating the output of the A / D converter 26 at the frequency of the phase modulation signal is averaged at 2τ time intervals. Alternatively, a demodulator configured with a computing unit that performs the operation may be used. Even in this case, as is clear from FIG. 22, a signal corresponding to the Sagnac phase difference φs can be extracted.

However, if the signal processing section 110b tries to extract a signal corresponding to the Sagnac phase difference φs using a demodulator constituted by an arithmetic unit for performing averaging processing, the following problem occurs.

In FIG. 4, when the filtering by DC remover 20 (for example, a band-pass filter) is insufficient, the DC component contained in the output of A / D converter 26 fluctuates (DC In the case where the component temporally changes with a certain slope), a sequential demodulator that adds the demodulation result of the first demodulator 112b and the demodulation result of the first demodulator 112b before τ time for every τ / m is used. If τ
The polarity of the DC component included in the signal corresponding to the Sagnac phase difference φs extracted at an interval of / m alternates, and the error due to the DC component is canceled.

For example, when m specifying the phase modulation frequency (m / 2τ) is 3, as shown in FIG. 33, at the input of A / D converter 26, the DC component is at at
(T is time). A
The / D converter 26 detects the fluctuation of the DC component by sampling every τ / 3. Therefore, the first
The signal input to the demodulator 112b has a,
2a, 3a, 4a, 5a, 6a,... At this time, assuming that the polarity of the reference signal B ′ input to the first reverter 112b is +, −, +, −..., The successive demodulator constituted by the τ delay unit 116b and the adder 118b Is -3a, 5τ at time 4τ / 3
+ 3a at / 3, -3a at 6τ / 3, + 3a at 7τ / 3,
..., + 3a and -3a are alternately output every τ / 3. That is, the fluctuation of the DC component at the input of the A / D converter 26 causes the polarity of the error to be alternately changed by the successive demodulator. Therefore, by successively integrating the output of the demodulator in the first computing unit 142 of the serodyne control unit 140, the fluctuation of the DC component can be canceled, and the Sagnac phase difference can be detected more accurately. And the performance of the gyro can be improved.

On the other hand, a demodulator constituted by an arithmetic unit averaging at a time interval of 2τ cannot cancel the fluctuation of the DC component.

For example, consider the case where m specifying the phase modulation frequency (m / 2τ) is 3.

As shown in FIG. 34, the DC component is at (t
A / D converter 26 detects the fluctuation of the DC component by sampling every τ / 3, and the signal input to the first demodulator 112b is: For each τ / 3, a, 2a, 3a, 4a, 5a,
6a,. . . . And change. At this time, the polarities of the reference signal B ′ input to the first demodulator 112b are +, −,
+, −,..., The demodulator output is −3a at time 4τ / 3, + 3a at 5τ / 3, −− at 6τ / 3.
3a. The non-sequential demodulator cannot output at any time every τ / 3, but outputs every τ. If the outputs are averaged at intervals of 2τ, an error of -3a / 2τ occurs. Similarly, an error of -3a / 2τ occurs between 2τ. That is, the fluctuation of the DC component at the input of the A / D converter 26 is not canceled, but becomes an error in the Sagnac phase difference detection, and causes a rotation angular velocity or a rotation angle error in the gyro output.

As described above, in the demodulator constituted by the arithmetic unit for averaging at the time interval of 2τ, the DC component included in the signal corresponding to the extracted Sagnac phase difference φs fluctuates and is canceled. Can not do.

The fluctuation of the DC component is caused by fluctuations in light intensity due to temperature, vibration, and the like, and mixing of low-frequency electromagnetic noise from the output of the light receiver 7 into the input section of the A / D converter 26. It is extremely difficult to completely suppress disturbance.

In the demodulator method of averaging at a time interval of 2τ (this method is also used in the optical fiber gyro described in the background art), a detection error occurs due to the fluctuation of the DC component, and the gyro output Causes a rotational angular velocity error or a rotational angle error.

If the above-described sequential demodulator is used in the signal processing section 110b of this embodiment (when m for specifying the frequency m / 2τ of the phase modulation signal is an odd number of 2 or more), the Sagnac due to the fluctuation of the DC component can be obtained. The detection error of the phase difference can be reduced, and the angular velocity or rotation angle output of the gyro can be improved.

Therefore, it can be said that the above-mentioned sequential demodulator is extremely effective demodulation means in an optical fiber gyro.

Note that the signal processing unit 110a shown in FIG.
(M for specifying the frequency (m / 2τ) of the phase modulation signal is 2
In the case of the above even number), the modulation and demodulation processing can also cancel the fluctuation of the DC component included in the signal corresponding to the detected Sagnac phase difference φs even in the demodulator that performs the averaging processing every 2τ. It is possible to improve the angular velocity or the rotation angle output of the gyro.

A demodulator that performs averaging every 2τ has the advantage that the operation speed is lower than that of a sequential demodulator. Therefore, in this embodiment,
An optical fiber gyro configured to perform the modulation and demodulation processing so that m specifying the frequency (m / 2τ) of the phase modulation signal is an even number of 2 or more can be said to be extremely effective.

The operation unit 122b includes the reference signal generator 300
The averaging process of the output signal of the second demodulator 114b is performed in the unit of time specified by the averaging time signal C generated in. This calculation result is a deviation signal of a modulation control system that is a closed loop including the modulation control unit 170b. This deviation signal is output to modulation control section 170b.

Here, in the phase modulation operation as shown in FIG. 2, it is assumed that the phase modulation by phase modulator 5 includes a modulation gain error. In this case, the modulation phase difference is as shown in FIG.

As shown in FIG. 23, simply by demodulating the output of A / D converter 26 at the frequency of the phase modulation signal by first demodulator 112b and averaging it by arithmetic unit 122b, the modulation gain A signal corresponding to the error (deviation signal of the modulation control system) cannot be extracted.

Therefore, in the present embodiment, a second demodulator 114b for demodulating at the frequency of the operating point switching signal of the phase modulation is provided, and the frequency and the phase of the phase modulated signal are compared with the output of the A / D converter 26. The demodulation is performed at the frequency of the modulation operation point switching signal, and the demodulation result is averaged. In this way, as shown in FIG. 23, a signal corresponding to the modulation gain error is extracted.

The deviation signal is generated by the modulation control section 170b when the modulation gain of the phase modulator 5 or the D / A converter 10 changes due to temperature change or aging. This is for adjusting the gain of the composite signal of the operating point switching signal of the phase modulation and the serrodyne signal. Therefore, the signal processing unit 110b and the modulation control unit 1
70b, and the servo loop including the interferometer 500
You don't have to be fast. In addition, since the response of the arithmetic unit 122b to the modulation control unit 170 may be several Hz to several hundred Hz,
The averaging time signal C may be of the order of msec. However, it is preferably an integer multiple of the frequency m / 2τ of the phase modulation signal.

As is clear from these descriptions, when m specifying the phase modulation frequency (m / 2τ) is an odd number of 2 or more, the Sagnac phase difference appears in the frequency band of m / 2τ. On the other hand, the modulation gain error is (m / 2τ) * (1 /
2m), and are equivalently represented by sideband spectra of (m-1) / 2τ and (m + 1) / 2τ.

For example, when m = 3, the Sagnac phase difference appears in the 3 / 2τ frequency band, and the modulation gain error appears in the 1 / τ and 2 / τ sideband spectral bands.

The band is limited by the sampling theorem.
When the Sagnac phase difference and the modulation gain error are detected by one A / D converter, some band limitation is considered.

A method of performing band limitation at a frequency of (m-1) / 2τ and performing sampling at a frequency of (m-1) / τ or more, a method of performing band limitation at a frequency of m / 2τ and performing a band limitation at a frequency of m / 2τ or more Method of sampling at frequency, (m + 1) / 2τ
, And a method of performing sampling at a frequency of (m + 1) / τ or more.

In these cases, as shown in FIG. 2, since the phase modulation is applied so that the directions of the optical spikes are always in the same direction, the dullness of the optical spikes after passing through the band limiting filter is obtained. The effect is always substantially constant near the sampling point, and the sampled optical spike components are canceled and eliminated through demodulation processing. As a result, it is possible to reduce the gyro scale factor error.

The effect of the dullness of the light spike is τ /
Since it is substantially symmetrical for each m, there is an advantage that a sampling point can be arbitrarily selected.

Furthermore, even when band limitation is not performed, the effects of optical spikes are substantially symmetrical for each τ / m, so that the effects of optical spikes can be offset and eliminated as described above. In addition, there is an advantage that the sampling point can be arbitrarily selected.

As described above, by performing the phase modulation so that the direction of the optical spike is always constant, the error due to the optical spike included in the detection optical signal becomes constant, and the error is canceled or removed through demodulation processing. Therefore, accurate modulation control can be performed, and the scale factor error of the gyro can be reduced.

Returning to FIG. 20, the description will be continued.

The modulation control section 170b performs a phase modulation signal of a rectangular wave for performing phase modulation so that the phase difference of the interference light is: m / 2τ, m is an odd number equal to or greater than 2 and θ: Is generated by the serrodyne control unit 140 and a rectangular wave phase modulation operating point switching signal for performing phase modulation so that the phase difference of is: frequency: 1 / 2τ amplitude: γ = kπ, k is an integer of 1 or more. The signal is combined with the cellodyne signal to generate a composite signal.

FIG. 24 is a block diagram of the modulation control unit 170 shown in FIG.
It is a schematic block diagram of b.

The reference value storage section 196 stores a reference output value necessary for causing a phase shift of θ (0 <θ ≦ π / 2 in the example shown in FIG. 2) and γ (in the example shown in FIG. 2, and a reference output value necessary for performing the phase shift of π).

The amplitude θ generator 202 reads out a reference output value necessary for performing a phase shift of θ stored in the reference value storage unit, and reads the reference output value from the optimum phase value transmitted from the light intensity calculation unit 210. The correction is performed based on the modulation amplitude value. Then, a constant output signal having the corrected value is generated.

The amplitude γ generator 208 reads a reference output value necessary for causing a phase shift of γ stored in the reference value storage unit, and generates a constant output signal having this value.

The phase modulation generator 204b outputs the reference signal B '(frequency: m / 2) generated by the reference signal generator 300.
τ, m is an odd number of 2 or more), and an amplitude θ generator 20
2 modulates the constant output signal generated.

As a result, a phase modulation signal of a rectangular wave for performing phase modulation is generated so that the phase difference of the interference light is a frequency: m / 2τ, where m is an odd number of more than 2 and the amplitude is θ.

This phase modulation generator 204b is formed of, for example, a multiplier.

Modulation generator 200b includes reference signal generator 3
The constant output signal generated by the amplitude γ generator 208 is modulated according to the reference signal A (frequency: 1 / 2τ) generated at 00.

As a result, a square wave phase modulation operating point switching signal for performing phase modulation so that the phase difference of the interference light becomes frequency: 1 / 2τ amplitude: γ is generated.

The adder 176b is connected to the phase modulation generator 204
b, the phase modulation operating point switching signal generated by the modulation generator 200b, and the serrodyne signal generated by the serrodyne control unit 140, and the phase modulation signal and the phase modulation operation are added. A composite signal of the point switching signal and the serrodyne signal is generated. At this time, the rising of the phase modulation signal and the rising of the reference signal A are synchronized so that both are added.

The first computing unit 172 includes a signal processor 110b
This is an integrator that integrates the deviation signal of the modulation control system sent from the control unit.

[0295] The second computing unit 174 is an amplifier or a low-pass filter. The gain or the filter constant is designed in accordance with the servo loop design of the modulation control system.

The multiplier 178 calculates the gain of the combined signal of the phase modulation signal, the phase modulation operating point switching signal, and the serrodyne signal output from the adder 176b.
The adjustment is made according to the deviation signal of the modulation control system output from 74. This result is output to the D / A converter 10.

The adder 198 outputs the output signal of the amplitude θ generator 202 (a value corresponding to the amplitude θ of the phase modulation by the phase modulation signal) and the output signal of the amplitude γ generator 208 (the operation signal for switching the operation point of the phase modulation). (A value corresponding to the amplitude γ of the phase modulation). This addition result is output to the light intensity calculation unit 210.

FIG. 25 is a block diagram showing modulation control section 170 shown in FIG.
3B shows the waveform of the output signal at each section when m for specifying the frequency of the phase modulation signal is set to 3. 25A shows the output signal waveform of the amplitude θ generator 202, FIG. 25B shows the output signal waveform of the phase modulation generator 204b, and FIG. 25C shows the output signal of the amplitude γ generator 208. FIG. 25D shows a signal waveform, FIG. 25D shows an output signal waveform of the modulation generator 200b, and FIG. 25E shows a composite waveform of the phase modulation signal and the phase modulation operating point switching signal.

As described above, the digital signal processor 1 in the case where m specifying the phase modulation frequency (m / 2τ) by the phase modulation signal is an odd number of 2 or more (for example, in the case of FIG. 2).
00b has been described.

In the present embodiment described above, a square wave phase modulation signal for performing phase modulation so that the phase difference of the interference light is frequency: m / 2τ, and m is an integer equal to or greater than 2 Combination of a square wave operating point switching signal and a serrodyne signal for performing phase modulation so that the phase difference of the interference light is: frequency: 1 / 2τ amplitude: γ = kπ, k is an integer of 1 or more The signal is generated such that the phase modulation signal and the phase modulation operating point switching signal are synchronized with the + θ phase shift by the phase modulation signal and the + γ phase shift by the phase modulation operating point switching signal.

[0301] Therefore, the error due to the optical spike included in the detection light signal is always constant, and the error is canceled or removed by the demodulation processing in the modulation control, so that more accurate modulation control can be performed. As a result, the scale factor error of the gyro can be reduced.

In this embodiment, the peak value P ₀ of the light intensity of the interference light is obtained, and a phase modulation signal is generated so as to have a phase modulation amplitude θ corresponding to the obtained value by referring to the table. ing. By doing so, it is possible to always suppress the influence of the random walk.

Further, according to the present embodiment, the following effects can be obtained.

A conventional digital optical fiber gyro is disclosed in, for example, Japanese Patent Application Laid-Open No.
No. 7931), the cellodyne signal (digital stepped ramp) is reset so as to perform a phase shift of -2π when the signal reaches 2π, and the signal between the cellodyne signal and the phase modulation signal is reset. The reset is also performed on the synthesized signal so that when the signal reaches 2π, a phase shift of −2π is performed.

Therefore, after the composite signal of the serrodyne signal and the phase modulation signal reaches 2π, the serrodyne signal becomes 2π.
The reset will be repeated until π is reached. In particular, when the input angular velocity to the optical fiber gyro is very low, the reset is repeated for a long time, and a dead zone called input lock-in which cannot measure the input angular velocity occurs. I will.

On the other hand, in the present embodiment, the reset is performed only for the serrodyne signal.
When the input angular velocity to the optical fiber gyro is very low, the reset is not repeatedly performed. Therefore, the lock-in phenomenon at the time of the low input angular velocity can be prevented.

[0307]

According to the present invention, an optical spike (a spike-like noise caused by a phase shift) included in interference light obtained by recombining two lights propagating in opposite directions through an optical fiber loop. , The direction of the optical spike at the end of the phase shift (the portion where the waveform is dull) is always constant. For this reason, the error due to the optical spike included in the detection light signal is always constant, so that the error can be canceled or removed by the demodulation processing in the modulation control, thereby enabling more accurate modulation control. As a result, the scale factor error of the gyro can be reduced.

Further, since the light intensity value of the interference light is detected and the phase modulation signal is generated so as to have the phase modulation amplitude corresponding to the detected value with reference to the table, the influence of the random walk is suppressed. be able to.

[Brief description of the drawings]

FIG. 1 is a diagram showing an embodiment of the present invention, in which a phase modulation signal is phase-modulated so that a phase difference of interference light becomes frequency: 1 / τ (m = 2) amplitude: 0 <θ ≦ π / 2 Using the square wave signal to be performed,
Modulation in the case of using a rectangular wave signal for performing phase modulation so that the phase difference of the interference light becomes: frequency: 1 / 2τ amplitude: γ = π (k = 1) as the operating point switching signal of the phase modulation FIG. 4 is a diagram for explaining a phase difference.

FIG. 2 shows an embodiment of the present invention, in which phase modulation is performed as a phase modulation signal such that the phase difference of the interference light is: frequency: 3 / 2τ (m = 3) amplitude: 0 <θ ≦ π / 2 Using the square wave signal to be performed,
Modulation in the case of using a rectangular wave signal for performing phase modulation so that the phase difference of the interference light becomes: frequency: 1 / 2τ amplitude: γ = π (k = 1) as the operating point switching signal of the phase modulation FIG. 4 is a diagram for explaining a phase difference.

FIG. 3 shows an embodiment of the present invention, in which phase modulation is performed as a phase modulation signal such that the phase difference of the interference light is: frequency: 2 / τ (m = 4) amplitude: π / 2 ≦ θ <π Using the square wave signal to be performed,
Modulation in the case of using a rectangular wave signal for performing phase modulation so that the phase difference of the interference light becomes: frequency: 1 / 2τ amplitude: γ = 2π (k = 2) as the operating point switching signal of the phase modulation FIG. 4 is a diagram for explaining a phase difference.

FIG. 4 is a schematic configuration diagram of a digital optical fiber gyro to which an embodiment of the present invention is applied.

FIG. 5 shows a phase modulation frequency (m / 2) by a phase modulation signal.
FIG. 9 is a schematic configuration diagram of the digital signal processor 100a when m for specifying τ) is an even number of 2 or more.

6 is a schematic configuration diagram of a signal processing unit 110a shown in FIG.

FIG. 7 is a diagram for explaining a case where an angular velocity is input to the optical fiber loop 6 and a Sagnac phase difference φs occurs in the phase modulation operation as shown in FIG. 1;

FIG. 8 is a diagram for explaining a case where an error is included in the modulation gain G in the phase modulation by the phase modulator 5 in the phase modulation operation as shown in FIG.

FIG. 9 shows an embodiment of the present invention, in which phase modulation is performed as a phase modulation signal such that the phase difference of the interference light is: frequency: 1 / τ (m = 2) amplitude: 0 <θ ≦ π / 2 Using the square wave signal to be performed,
Detection when a square wave signal for performing phase modulation is used as the phase modulation operating point switching signal so that the phase difference of the interference light is: frequency: 1 / 2τ amplitude: γ = π (k = 1) A when no band limitation is applied to light intensity
FIG. 4 is a diagram for explaining a / D converter input signal waveform.

FIG. 10 is a schematic configuration diagram of a serrodyne control unit 140 shown in FIG. 5;

11 is a diagram showing an operation flow of the comparator 150 shown in FIG. 10 when the first threshold is set to + 2π and the second threshold is set to −2π.

12 is a diagram showing an operation flow of the comparator 150 shown in FIG. 10 when the first threshold value is set to + 2π and the second threshold value is set to 0.

13 is a diagram showing an operation flow of the comparator 150 shown in FIG. 10 when the first threshold is set to + π and the second threshold is set to -π.

FIG. 14 is a diagram showing a waveform of a serrodyne signal generated by a third computing unit 146 shown in FIG.

15 is a schematic configuration diagram of a modulation control unit 170a shown in FIG.

FIG. 16 shows a modulation control unit 170a shown in FIG.
FIG. 9 is a diagram illustrating waveforms of output signals at respective units when m that specifies the frequency of the phase modulation signal is 2.

17 is a schematic configuration diagram of a light intensity calculation unit 210 shown in FIG.

FIG. 18 shows the relative intensity noise RIN of the light source 1 as −115 [d]
B / Hz], and the peak value P ₀ of the light intensity of the interference light is 2 to 50
FIG. 9 is a diagram showing a random walk value-phase modulation amplitude characteristic when μW is set.

19 is a schematic configuration diagram of a gyro output operation unit 240 shown in FIG.

FIG. 20 shows a phase modulation frequency (m / 2) by a phase modulation signal.
FIG. 9 is a schematic configuration diagram of the digital signal processor 100b when m for specifying τ) is an odd number of 2 or more.

21 is a schematic configuration diagram of a signal processing unit 110b shown in FIG.

FIG. 22 is a diagram for explaining a case where an angular velocity is input to the optical fiber loop 6 and a Sagnac phase difference φs occurs in the phase modulation operation as shown in FIG.

FIG. 23 is a diagram for explaining a case where an error is included in the modulation gain G in the phase modulation by the phase modulator 5 in the phase modulation operation as shown in FIG.

24 is a schematic configuration diagram of a modulation control unit 170b shown in FIG.

FIG. 25 is a diagram illustrating a modulation control unit 170b shown in FIG.
FIG. 7 is a diagram illustrating waveforms of output signals at respective units when m that specifies the frequency of the phase modulation signal is set to 3;

FIG. 26 is a diagram for explaining a modulation phase difference by a conventional gyro.

FIG. 27 is a diagram for explaining a modulation phase difference by a conventional gyro.

FIG. 28 is a diagram for explaining a modulation phase difference by a conventional gyro.

FIG. 29 is a diagram showing an A / D converter input signal waveform by a conventional gyro.

FIG. 30 is a diagram showing an input signal waveform of an A / D converter using a conventional gyro.

FIG. 31 is a diagram illustrating an embodiment of the present invention, in which phase modulation is performed as a phase modulation signal such that the phase difference of the interference light is: frequency: 1 / τ (m = 2) amplitude: 0 <θ ≦ π / 2 Using the square wave signal to be performed,
Detection when a square wave signal for performing phase modulation is used as the phase modulation operating point switching signal so that the phase difference of the interference light is: frequency: 1 / 2τ amplitude: γ = π (k = 1) FIG. 7 is a diagram for explaining an A / D converter input signal waveform when a band is limited to 1 / 2τ with respect to light intensity.

FIG. 32 is a diagram showing an embodiment of the present invention, in which phase modulation is performed as a phase modulation signal such that the phase difference of the interference light is: frequency: 1 / τ (m = 2) amplitude: 0 <θ ≦ π / 2 Using the square wave signal to be performed,
Detection when a square wave signal for performing phase modulation is used as the phase modulation operating point switching signal so that the phase difference of the interference light is: frequency: 1 / 2τ amplitude: γ = π (k = 1) FIG. 5 is a diagram for explaining an A / D converter input signal waveform when a band is limited to 1 / τ with respect to light intensity.

FIG. 33 is a diagram illustrating an output of a sequential demodulator system when the frequency of a phase modulation signal is 3 / 2τ (m = 3) and the DC component of the input of the A / D converter is fluctuating in one embodiment of the present invention; It is a figure for explaining a characteristic.

FIG. 34 is a diagram illustrating a non-sequential demodulator method in a case where the frequency of the phase modulation signal is 3 / 2τ (m = 3) and the DC component of the input of the A / D converter fluctuates in one embodiment of the present invention. FIG. 4 is a diagram for explaining output characteristics.

[Explanation of symbols]

REFERENCE SIGNS LIST 1 light source 2, 4 coupler 3 polarizer 5 phase modulator 6 optical fiber loop 7 light receiver 8 current / voltage (I / V) converter 9, 22 amplifier 10 D / A converter 11 driver 20 DC remover 26, 32 A / D converters 100, 100a, 100b Digital signal processors 110a, 110b Signal processing units 112a, 112b First demodulators 114a, 114b Second demodulators 116a, 116b τ delayers 118a, 118b, 176a, 176b, 198, 2
06 Adders 122a, 122b Operators 134, 136, 178 Multipliers 140 Serrodyne control units 142, 172, 212, 242 First operators 144, 174, 214, 246 Second operators 146, 218 Third operators 148 Reference Value storage unit 150 Comparator 170a, 170b Modulation control unit 200a, 200b Modulation generator 202 Amplitude θ generator 204a, 204b Phase modulation generator 208 Amplitude γ generator 210 Light intensity calculator 220 Fourth calculator 222 Fifth calculator 240 Gyro output operation unit 244 Register 300 Reference signal generator 500 Interferometer

 ──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Kanji Yamamoto 2-16-46 Minami Kamata, Ota-ku, Tokyo Inside Tokimec Co., Ltd.

Claims (23)

[Claims] An optical fiber gyro for detecting a rotation angular velocity or a rotation angle according to a Sagnac phase difference of interference light obtained by recombining two lights propagating in an optical fiber loop in opposite directions, A signal generation unit that generates a composite signal of a phase modulation signal, an operation point switching signal for phase modulation, and a serrodyne signal for canceling a Sagnac phase difference, and according to the composite signal generated by the signal generation unit, A phase modulator for phase-modulating each of two lights propagating through the optical fiber loop, wherein the signal generation unit determines a phase difference of the interference light when a propagation time of the light in the optical fiber loop is τ. Where: frequency: m / 2τ, m is an integer greater than or equal to 2 Amplitude: θ A phase modulation signal of a rectangular wave for performing phase modulation so as to be: τ amplitude: γ = kπ, where k is an integer of 1 or more, an operating point switching signal for rectangular wave phase modulation for performing phase modulation, and duration of each step: τ / n, where n is an integer of 1 or more A stepped serodyne signal and a combined signal of the phase modulation signal and the + γ phase shift by the phase modulation operating point switching signal are generated in synchronization with each other. Fiber gyro. 2. The optical fiber gyro according to claim 1, wherein k is an odd number: 0 <θ ≦ π / 2, and k is an even number: π / 2 ≦ θ <π. 3. An optical fiber gyro according to claim 1, wherein k is 1. 4. The optical fiber gyro according to claim 1, wherein m is an even number of 2 or more. 5. The optical fiber gyro according to claim 4, wherein said signal generating means includes a first constant output signal generating means for generating a signal necessary for performing a phase shift of θ as a constant output signal. A first modulation unit for modulating the constant output signal generated by the first constant output signal generation unit at a frequency of m / 2τ to generate the phase modulation signal, and performing a phase shift of γ. Second constant output signal generating means for generating a signal necessary for the constant output signal, a phase modulation signal generated by the first modulation means and a constant output generated by the second constant output signal generating means First addition means for adding a signal to generate a signal; and modulating the signal generated by the first addition means at a frequency of 1 / 2τ, thereby obtaining the phase modulation signal and the phase modulation signal. Converting the synthesized signal with the operating point switching signal into the phase modulated signal; Second modulating means for generating a phase shift of + θ and a phase shift of + γ by the operating point switching signal of the phase modulation in synchronism with each other; a composite signal generated by the second modulating means and the cellodyne. An optical fiber gyro, comprising: a second adding unit that adds a signal to generate a composite signal of the phase modulation signal, the operating point switching signal of the phase modulation, and the serrodyne signal. 6. An optical fiber gyro according to claim 5, wherein said light receiving means detects the light intensity of said interference light and converts it into an electric signal, and demodulates the output of said light receiving means at a frequency of m / 2τ. A first demodulation unit, a second demodulation unit for demodulating an output of the first demodulation unit at a frequency of 1 / 2τ, a delay unit for delaying an output of the second demodulation unit by a time τ, Third adding means for adding the output of the second demodulating means and the output of the delay means to detect a signal corresponding to the Sagnac phase difference, and wherein the signal generating means comprises the third An optical fiber gyro, further comprising: a serrodyne signal generating means for generating the serrodyne signal in accordance with a signal corresponding to the Sagnac phase difference output from the adding means. 7. An optical fiber gyro according to claim 6, wherein an arithmetic means for detecting a signal corresponding to a modulation gain error by averaging an output of said first demodulation means at predetermined time intervals. The signal generation unit further includes a modulation gain control unit that controls a gain of a composite signal output from the second addition unit according to a signal corresponding to a modulation gain error output from the calculation unit. An optical fiber gyro characterized by the above. 8. The optical fiber gyro according to claim 6, wherein a gyro output calculating means for calculating a rotational angular velocity or a rotational angle based on a signal corresponding to the Sagnac phase difference output from the third adding means. A light intensity calculating unit for obtaining a light intensity value of the interference light from an output of the light receiving unit; and a table showing a correspondence relationship between the light intensity value and a correction coefficient for correcting a scale factor error of the gyro output. Means, and the gyro output calculating means corrects the calculation result according to a correction coefficient corresponding to the light intensity value of the interference light obtained by the light intensity calculating means, which is shown in the table. An optical fiber gyro characterized by outputting. 9. The optical fiber gyro according to claim 6, wherein: a light intensity calculating means for obtaining a light intensity value of the interference light from an output of the light receiving means; According to the difference between the light intensity value and a predetermined reference value, the AGC (Automatic Ga
AGC signal generation means for generating a signal, and a loop gain control means for controlling a gain of a signal output from the light receiving means in accordance with the AGC signal generated by the AGC signal generation means. An optical fiber gyro characterized by the following. 10. An optical fiber gyro according to claim 1, wherein m is an odd number of 2 or more. 11. The optical fiber gyro according to claim 10, wherein said signal generating means includes: first constant output signal generating means for generating a signal necessary for performing a phase shift of θ as a constant output signal; A first modulating means for modulating the constant output signal generated by the first constant output signal generating means at a frequency of m / 2τ to generate the phase modulated signal, and performing a phase shift of γ. A second constant output signal generating means for generating a signal necessary for the constant output signal, and modulating the constant output signal generated by the second constant output signal generating means at a frequency of 1 / 2τ, A second modulation unit that generates an operation point switching signal of phase modulation, a phase modulation signal generated by the first modulation unit, an operation point switching signal of phase modulation generated by the second modulation unit,
First adding means for adding the serrodyne signal so that the + θ phase shift by the phase modulation signal and the + γ phase shift by the phase modulation operating point switching signal are synchronized to generate the composite signal And an optical fiber gyro, comprising: 12. An optical fiber gyro according to claim 11, wherein a light receiving means for detecting a light intensity of the interference light and converting the light intensity into an electric signal, and demodulating an output of the light receiving means at a frequency of m / 2τ. First demodulation means, delay means for delaying the output of the first demodulation means by time τ, adding the output of the first demodulation means and the output of the delay means, to the Sagnac phase difference Second adding means for detecting a corresponding signal, wherein the signal generating means generates the serrodyne signal according to a signal corresponding to the Sagnac phase difference output from the second adding means. An optical fiber gyro further comprising signal generation means. 13. An optical fiber gyro according to claim 12, wherein said second demodulation means demodulates an output of said first demodulation means at a frequency of 1 / 2τ, and outputs an output of said second demodulation means. Calculating means for detecting a signal corresponding to the modulation gain error by averaging at a predetermined time interval, wherein the signal generating means converts the signal corresponding to the modulation gain error output from the calculating means to Accordingly, the optical fiber gyro further comprises a modulation gain control means for controlling a gain of the composite signal output from the first addition means. 14. An optical fiber gyro according to claim 12, wherein a gyro output calculating means for calculating a rotation angular velocity or a rotation angle based on a signal corresponding to the Sagnac phase difference outputted from said second addition means. A light intensity calculating unit for obtaining a light intensity value of the interference light from an output of the light receiving unit; and a table showing a correspondence relationship between the light intensity value and a correction coefficient for correcting a scale factor error of the gyro output. Means, and the gyro output calculating means corrects the calculation result according to a correction coefficient corresponding to the light intensity value of the interference light obtained by the light intensity calculating means, which is shown in the table. An optical fiber gyro characterized by outputting. 15. The optical fiber gyro according to claim 12, wherein: a light intensity calculating means for obtaining a light intensity value of the interference light from an output of the light receiving means; According to the difference between the light intensity value and a predetermined reference value, the AGC (Automatic Ga
AGC signal generation means for generating a signal, and a loop gain control means for controlling a gain of a signal output from the light receiving means in accordance with the AGC signal generated by the AGC signal generation means. An optical fiber gyro characterized by the following. 16. An optical fiber gyro according to claim 1, 2 or 3, wherein a light receiving means for detecting a light intensity of the interference light and converting the light intensity into an electric signal; Light intensity calculating means for obtaining a light intensity value, wherein the signal generating means generates the phase modulation signal such that θ changes according to the light intensity value of the interference light obtained by the light intensity calculating means. An optical fiber gyro characterized by: 17. The optical fiber gyro according to claim 16, further comprising means for storing a table indicating a correspondence relationship between light intensity and θ, wherein said signal generation means includes: An optical fiber gyro, wherein the phase modulation signal is generated so as to be θ corresponding to the light intensity value of the interference light obtained by the light intensity calculating means. 18. The optical fiber gyro according to claim 8, 9, 14, 15, 16, or 17, wherein the light intensity calculating means is configured to determine the interference based on an output of the light receiving means and θ and γ. An optical fiber gyro, wherein a peak value of a light intensity of the interference light is obtained as a light intensity value of the light. 19. An optical fiber gyro according to claim 18, wherein said light intensity calculating means sets a peak value of light intensity of said interference light to P ₀ and an output of said light receiving means to V _out . An optical fiber gyro, wherein a peak value of the light intensity of the interference light is obtained from P ₀ = 2V _out / {1 + cos (γ ± θ)}. 20. The optical fiber gyro according to claim 1, wherein said signal generating means includes: a serrodyne signal generating means; and a reset means for resetting generation of a serrodyne signal by the serrodyne signal generating means. The resetting means resets the serrodyne signal generation means so that when the accumulation result of the modulation phase difference by the serrodyne signal reaches a first threshold, the phase difference of the interference light becomes a phase shift of -2π. And when the accumulation result of the modulation phase difference due to the serrodyne signal reaches a second threshold value that is 2π lower than the first threshold value, the serrodyne signal generation is performed so that the phase difference of the interference light becomes a phase shift of + 2π. An optical fiber gyro characterized by resetting means. 21. An optical fiber gyro according to claim 20, wherein said first threshold value is + π and said second threshold value is -π. 22. The optical fiber gyro according to claim 20, wherein the first threshold is + 2π and the second threshold is zero. 23. The optical fiber gyro according to claim 1, 2 or 3, wherein the signal generating means is a serrodyne signal generating means, and a serrodyne before being combined with the phase modulation signal and the operating point switching signal. An optical fiber gyro, further comprising: reset means for resetting generation of a serrodyne signal by the serrodyne signal generating means when the signal reaches a threshold value.