JPH0422930A - Optical identification reproducer - Google Patents

Abstract

PURPOSE:To obtain this optical identification reproducer capable of driving at a super high speed by providing the reproducer with an optical amplifier for inputting an optical signal pulse and an optical clock pulse synchronized with the optical signal pulse and amplifying only the part of the optical signal pulse which is hourly superposed to the optical clock pulse and an optical threshold element for outputting light with non-linear intensity. CONSTITUTION:The phases of the optical signal pulse string 1 and the optical clock pulse 2 are adjusted so that the peak of the pulse 2 is aligned on the average peak position of inputted signal optical pulses and both pulses 1, 2 are multiplexed by a multiplexer 5 and inputted to the light amplifier 7. The amplifier 7 amplifies only the part of the optical signal pulses which is hourly superposed to the optical clock pulse and the amplified optical signal pulse is separated from the optical clock pulse and natural discharge light by a demultiplexer 6. The amplified optical signal pulse is inputted and identified by the optical threshold element 8 indicating a non-linear response whose light output intensity or light output energy is changed in stages and outputted as an optical signal pulse string 4 having no amplitude noise. Consequently, equalizing amplification, retiming and identification reproduction can be executed at a very high speed.

Description

Detailed Description of the Invention (Industrial Application Field) The present invention relates to an optical communication device, and in particular, performs equalization amplification in the optical domain on a weak time-division multiplexed optical signal pulse train propagated through a transmission path. The present invention relates to an all-optical optical identification regenerator that performs , retiming, and identification regeneration.

(Prior art) In current optical communication equipment, optical signals actually flow only in the transmission path, and in repeaters, the optical signals are converted into electrical signals, and electronic circuits convert the optical signals into electrical signals. ,
After being subjected to equalization amplification, retiming, and identification/regeneration processing, it is converted back into an optical signal and transmitted. Since optical/electrical conversion and electrical/optical conversion are performed in this way, the device configuration has become extremely complicated. Furthermore, in processing using electronic circuits, there is a theoretical processing speed limit at a bit rate of 10 to several tens of Gb/s. In order to solve the problems of complicating the device configuration and limiting processing speed due to the intervention of electrical processing as described above, it is necessary to convert the optical signal as it is at high speed without converting it to an electrical signal. An optical repeater is required to handle this.

Such all-optical optical repeaters have conventionally utilized stimulated Raman scattering or stimulated Brillouin scattering in fibers, or stimulated emission in semiconductor lasers or erbium-doped fibers to recover light that has attenuated as it propagates through optical fibers. There are optical direct amplifiers that directly amplify signals as they are optical signals, and all of them other than stimulated Brilliant amplifiers have a response speed of sub-picoseconds.

Regarding stimulated Raman amplification, Y. Aoki: “Pr.
operations of fiber Raman a
mplifiers and their apps
ability of digital optic
al com+unication systems,
'IEEE J, Lighthave Techno
l,, LT-6 (1988) 1225-1239. , for induced Brilliant amplification, C1G, Atkins
et al.: Application of
Brillouinamplification i
n coherent optical tra
nsmission, 'Electron, L
ett, 22 (1986) 556-558. , regarding semiconductor laser amplifiers, T. Mukai et a.
l, :'0ptical amplificatio
n by semiconductor laser
s, ” Sem1 conductors and S
emimetals, 22 (Academic Pr.
ess, 1985) 265-319. , for erbium-doped fiber amplifiers, D, N, Payne e
tal:Rare-earth-doped 1a
sers and amplifiers.

EC0C'88. L (198B) 49-53, respectively.

(Problems to be Solved by the Invention) Although the above-mentioned optical direct amplifier has high-speed response, it is basically a linear amplifier, and during long-distance multi-stage relaying, there are problems of waveform deterioration due to accumulation of noise and jitter and dispersion. It arises. In order to solve these problems, optical regenerative amplification that performs equalization amplification, retiming, and discrimination regeneration in the optical domain is essential. In view of the above points, it is an object of the present invention to provide an all-optical optical identification regenerator that can operate at ultra high speed without being subject to electronic circuit speed limitations.

(Means for Solving the Problems) The optical identification regenerator of the present invention inputs an optical signal pulse input from an optical transmission line and an optical clock pulse synchronized with the optical signal pulse. an optical amplifier that amplifies only the overlapping portion and outputs the amplified light, and an optical amplifier that inputs the optical output of this optical amplifier and outputs light with an optical intensity that is nonlinear with respect to the optical intensity of this optical input. and a light threshold element.

(Function) The optical intensity modulated binary digital optical signal pulse train propagates through the optical transmission line while undergoing optical intensity attenuation and waveform distortion.
It is input to the optical discriminator of the present invention.

Further, the optical clock pulse, which is generated by an optical timing extraction circuit not included in the present invention, is synchronized with the input optical signal pulse train, and has a pulse time width sufficiently narrower than the input signal pulse.
The phase of the pulse is adjusted so that the peak of the pulse corresponds to the average peak position of the input signal light pulse, and the pulse is input to the optical discriminator of the present invention. The optical signal pulse train and the optical clock pulse are first input to an optical amplifier, where only the portion of the optical signal pulse that temporally overlaps with the optical clock pulse is amplified and output from the optical amplifier. As a result, the jitter of the optical signal pulse is absorbed (retiming) and the waveform is shaped (equalization amplification).

The amplified optical signal pulses are affected by amplitude noise originally inherent in each input optical signal pulse, as well as amplitude irregularities caused by the deviation between the peak position of the optical signal and the peak position of the optical clock pulse, and noise caused by the optical amplifier. It has spontaneous emission optical noise. These are identified by inputting the optical output intensity or optical output energy to an optical threshold element that exhibits a step-like nonlinear response to the optical input intensity or optical input energy (discriminative reproduction), and are free from amplitude noise. It is output as an optical signal pulse train.

(Embodiment) FIG. 1(a) is a diagram showing the configuration of an embodiment of the optical discrimination regenerator of the present invention, in which 1 is a binary digital input optical signal pulse train that is light intensity modulated; While propagating through the transmission path, the light is subject to attenuation of light intensity and waveform distortion. 2 is an optical clock pulse generated by an optical timing extraction circuit not included in the present invention, synchronized with the input optical signal pulse train 1, and having a pulse time width sufficiently narrower than the input signal pulse 1. These light intensity waveforms are shown in Fig. 1 et al. In addition to amplitude noise, each input optical signal pulse has a random shift (jitter) between the peak position of the optical signal and the peak position of the optical clock pulse. The optical signal pulse train and the optical clock pulse are combined in phase by a multiplexer 5 and input to an optical amplifier 7, with the phases adjusted so that the peak of the optical clock pulse is at the average peak position of the input signal optical pulse. Ru. Here, only the portion of the optical signal pulse that temporally overlaps with the optical clock pulse is amplified, and the amplified optical signal pulse is separated from the optical clock pulse and the spontaneous emission light by the optical demultiplexer 6.

As a result, the jitter of the optical signal pulse is absorbed (retiming) and the waveform is shaped (equalization amplification).

As shown in Fig. 1('b), the amplified optical signal pulse is affected by the amplitude noise between the peak position of the optical signal and the peak position of the optical clock pulse, in addition to the original amplitude noise of each input optical signal pulse. There is amplitude irregularity due to the deviation of the optical amplifier, and spontaneous emission optical noise of the optical amplifier. These are identified by inputting the optical output intensity or optical output energy to the optical threshold element 8 which shows a step-like nonlinear response to the optical input intensity or optical input energy (discriminative reproduction), and there is no amplitude noise. It is output as an optical signal pulse train 4.

If the optical intensity of the input optical signal pulse is small and the gain of the optical amplifier alone is insufficient, an optical amplifier such as a semiconductor laser amplifier or a rare earth-doped fiber amplifier can be placed in front of the optical multiplexer 5. Further, if the input light intensity or output light intensity of the optical threshold element 8 is less than a predetermined value, a similar optical amplifier can be placed before or after the optical threshold element 8.

Next, the optical amplifier and the optical threshold element, which are the constituent elements of the present invention, will be explained in detail.

1-Man - Figure 2(a) shows an example of the configuration of an optical amplifier using stimulated Raman scattering that can be used in the present invention, where numeral 2 represents an input optical signal pulse train, 2 represents an optical clock pulse, and 9 represents a dichroic mirror. , 10 is a single mode optical fiber, and 11 is an optical filter. Figure 2(b) shows each point A in Figure 2(a).
, B and C are shown.

Raman scattering is when a material is irradiated with monochromatic light (pump light), the interaction between the optical phonons of the material and the irradiated monochromatic light causes light (Stokes light) is mixed into the scattered light. When an amorphous material such as an optical fiber is used as the material, the spectrum width of Stokes light is wider and its peak intensity is lower than when a crystal is used. Taking GeO□+'zos+SrO□ as an example of the oxide glass, the Raman gain coefficient is shown in FIG. SiO□, which is used as a material for ordinary optical fibers, has a gain in terms of wavelength shift (Stokes shift amount) in the range of 50 to 600 cm-', and the Stokes shift amount is 440 C! O
-', it has a gain of I This is a phenomenon in which light is generated.Furthermore, when light with a wavelength within the Raman gain band is incident at the same time as pump light, gain is obtained from the pump light and the incident light is amplified.

In the case of a silica-based fiber, use the above value as the Raman gain coefficient and define the effective core cross-section as lXl0-”CID2,
If the pump light power is 5- and the fiber length is 11V, then 2
A gain of 1.7 dB will be obtained. Stimulated Raman scattering has a wide gain band, high amplification factor, and sub-picosecond response speed. Furthermore, since the gain can be changed by the pump light, gating for a very short period of pico to sub-pico seconds is possible using an extremely short pulse of the pump light.

In FIG. 2, an input optical signal pulse train 1 and an optical clock pulse 2 which is synchronized therewith and whose pulse time width is sufficiently narrower than the input optical signal pulse are multiplexed by a dichroic mirror 9,
input into a single moat optical fiber 10. Here, it is assumed that the optical clock pulse is used as pump light and the wavelength of the signal optical pulse is within the Raman gain band. Stimulated Raman scattering does not have a clear dark value for the pump light intensity, but the pump light intensity is set to a value such that strong Stokes light is not generated by the pump light alone in the absence of a signal light pulse (
■) in Figure 2(b). The weak Stokes light at this time is called spontaneous emission light, following the example of a laser with a normal resonator structure.

The phases of such extremely short optical clock pulses and optical signal pulse trains are adjusted so that the peak of the optical clock pulses is at the average peak position of the input optical signal pulses, and then input into an optical fiber. When the optical signal pulse is "1", only the portion of the optical signal pulse that temporally overlaps with the optical clock pulse is amplified. On the other hand, the optical signal pulse is
When the angle is 0°, no amplification occurs and only weak spontaneous emission light appears.The amplified signal light pulse is shown in Figure 2(b).
The clock light pulse and the spontaneous emission light are separated from each other by an optical filter 11 having a transmission band as shown in (2) and output.

Normally, when the pump light wavelength is in the anomalous dispersion region, modulation instability, which will be described later, is more likely to occur with a weaker pump light intensity than stimulated Raman scattering. In order to avoid this, a single mode fiber having normal dispersion at the pump light wavelength may be used.

Furthermore, in order to eliminate walk-off caused by the difference in group velocity between the pump light and the signal light, the fiber is designed so that the zero dispersion wavelength is between the pump light wavelength and the signal light wavelength, and the group delay between the pump light and the signal light is reduced. All you have to do is eliminate it.

In addition, when the signal light wavelength is in the anomalous dispersion region, the optical signal undergoes Raman amplification due to chirping and anomalous dispersion caused by the optical power effect, in which the refractive index of the material changes depending on the light intensity, and its intensity increases. becomes stronger and higher-order solitons are generated, compression of the pulse occurs (soliton compression).

This can be used to shape the pulse.

Furthermore, instead of the dichroic mirror 9 or the optical filter 11, an optical fiber component having the same characteristics as these may be used.

Modulation Insufficiency Amplifier Next, an optical amplifier based on modulation instability that can be used in the present invention will be described. FIG. 4(a) shows a configuration example of an optical amplifier that takes advantage of modulation instability that can be used in the present invention, where 1 is an optical signal optical pulse train, 2 is a clock optical pulse, 9 is a dichroic mirror, and IO is a clock. A single mode optical fiber having anomalous dispersion at the wavelength of the optical pulse, 11 is an optical filter. Further, FIG. 4(b) shows the dispersion characteristics of the single mode optical fiber 10 and FIG. 4(a).
Each point A inside.

The light spectra in B and C are shown.

Modulation instability is a phenomenon in which when a continuous wave of monochromatic light is input into an optical fiber, its amplitude is not stable, and noise with a small amplitude grows and becomes a soliton. In the frequency domain, it is represented by parametric four-photon mixing between pump light (angular frequency ωP, wave number kp), Stokes light (angular frequency ωS, wave number ks), and anti-Stokes light (angular frequency ωa1 wave number ka). These lights must satisfy the following relationship.

2ωP=ωS+ωa(1) 2kp-ks-ka-0(2) Equation (2) is a phase matching condition, and when the pump light wavelength is in the anomalous dispersion region of the optical fiber used, the Kerr effect due to the pump light There is a Stokes wavelength where the nonlinear refractive index change and linear dispersion are balanced and the phase matching condition is satisfied. When the pump light is in the anomalous dispersion region as described above, gain is obtained from the pump light through parametric four-photon mixing, and sidebands grow. The wavelength dependence of the sideband gain coefficient is expressed by the following equation.

g(Ω), ni'Ω" [2ωn, E"/(ck'Ω")-
11”!/2 (3) Here, Ω is the angular frequency difference with the pump light, k# is the second derivative of the wave number at the pump light wavelength, ω is the angular frequency of the pump light, and nf is the nonlinear force constant (
1,22X 10-”I12/v, E is the electric field amplitude, C
is the speed of light in vacuum.

This situation is shown in FIG. 5, with the horizontal axis plotting the difference from the angular frequency of the pump light. The gain coefficient is maximum at the angular frequency Ω□8 where the phase matching condition is completely satisfied, which is Ωm, x = (ωnJ2/ck'') 1/2
It is given by (4). k'' is k using group velocity dispersion,
=Dλ2/2πC, Ω□8, and therefore the gain band becomes larger as the group velocity dispersion becomes smaller and the pump light intensity becomes stronger. The maximum gain coefficient g□ at that time,
is gmjx = nzωE2/2c (
5) The stronger the pump light intensity, the larger the output. - As an example, the pump light intensity is set to 3.2, and the nonlinear force constant is set to 3.2.
Xl0-16cm28, the effective core cross-sectional area of the fiber is I
Xl0-'C'm" and the wavelength is 1.55 μm, the maximum gain coefficient g11..l is 3.2 X 10-"/
m. If the fiber length is 2 km, a gain of 28 dB can be obtained.

Based on modulation instability (parametric link amplification has a wide gain band, high amplification factor, and picosecond response speed. Furthermore, since the gain can be changed by pump light, it is possible to use extremely short pulses of pump light. Therefore, very short gating time of picoseconds is possible.

In FIG. 4, an input optical signal pulse train 1 and an optical clock pulse 2 that is synchronized therewith and whose pulse time width is sufficiently narrower than the input signal pulse are multiplexed by a gyroic mirror 9 and connected to a single mode optical fiber lo. incident. Here, it is assumed that the optical clock pulse is used as pump light, and the wavelength of the signal optical pulse is within the modulation unstable gain band. The pump light intensity is set to a value such that strong Stokes light and anti-Stokes light are not generated by the pump light alone in the absence of signal light pulses (■ in Figure 4). The weak Stokes light and anti-Stokes light at this time are called spontaneous emission light, following the example of a laser with a normal resonator structure.

The phases of such extremely short optical clock pulses and optical signal pulse trains are adjusted so that the peak of the optical clock pulses is at the average peak position of the input optical signal pulses, and then input into an optical fiber. When the optical signal pulse is 1", only the part of the optical signal pulse that temporally overlaps with the optical clock optical pulse is amplified, and parametric link amplification changes the spectrum of the incident signal light at the frequency of the pump light. Amplified signal light also appears at the same frequency (■ in FIG. 4(b)).On the other hand, when the optical signal pulse is "0°", no amplification occurs and only weak spontaneous emission light appears.

The two amplified signal light pulses are separated into one by an optical filter 11 having a transmission band a or a transmission band 2 as shown in (2) in FIG. 4(b), and are output. The wavelength of the incident signal light may be within the gain band of either the Stokes light side or the anti-Stokes light side,
Furthermore, any one of the amplified optical signal pulses may be taken out as an output.

Furthermore, instead of the guichroic mirror 9 or the optical filter 11, an optical fiber component having the same characteristics as these may be used.

FIG. 6(a) shows the nonlinear Sa that can be used in the present invention.
It is a diagram showing a configuration example of an optical threshold element based on a gnac interferometer, in which 18 is an optical isolator, 19 is an optical coupler with a power branching ratio of not 0.5, and 20 is an optical fiber loop. After passing through the isolator 18, the human power light 13 is split into two by an optical coupler, and then becomes a clockwise light 14 and a counterclockwise light 15, which propagate in the optical fiber loop 20 in opposite directions. After that, the clockwise light 14 and the counterclockwise light 15 return to the optical coupler 19 again, and some of them are transmitted to the boat P.
The light 16 is output from + and returns in the direction from which it came, and the rest becomes light 17 output from the boat P2.

If the dispersion ratio of the optical coupler 19 is not 0.5, the light intensities of the clockwise light 14 and the counterclockwise light 15 will be different.
The magnitude of the self-phase modulation caused by the optical force-effect is different, which results in a phase difference Δφ between the two after loop propagation. The optical power It(t) of the light 17 output from the boat Pz changes depending on the phase difference Δφ as given by the following equation.

Iy(t)=1°(t) (1-2k(1-k)(1+
cosΔφ) ) (6)Δφ−2πnzL(1-2
k) Io(t) /λA (7) Here, Io(t) is the input optical power waveform, n2 is the nonlinear constant, L is the loop length,
k is the power branching ratio of the optical coupler, λ is the wavelength, and A is the effective core cross-sectional area.

In this situation, the horizontal axis is the optical power of the input light, and the vertical axis is the boat P2.
The optical power of the light 17 outputted from is shown in FIG. Since self-phase modulation due to the optical power effect has a response speed on the order of femtoseconds, the optical F output from the boat pz changes depending on the instantaneous optical power of the incident light. As is clear from this figure, the optical power of the light 17 output from the boat Pz shows a step-like nonlinear response to the input optical power, so it can be used as an ultra-high-speed dark value element. The light 16 output from the boat P is the isolator 1
8 is removed.

In order to make this response steeper, it is necessary to
), a large number of the nonlinear Sagnac interferometers just described may be connected in series. The response at this time is determined by taking the number of columns of the nonlinear Sagnac interferometer as a parameter, n.
It is shown in FIG. 7('b). It can be seen that the response becomes steeper as the number of columns increases.

The output light of the boat Pz changes depending on the instantaneous optical power of the input light. Since the optical waveform of normal input light is not rectangular,
By passing through this optical threshold element, the output light from the boat Pg undergoes waveform distortion as well as dark value processing. In order to avoid this, an optical fiber having anomalous dispersion at the input light wavelength may be used for the optical fiber loop 20, and the clockwise light 14 and the counterclockwise light 15 may be propagated as optical solitons.

An optical soliton is a special optical pulse that propagates over long distances in an optical fiber without waveform distortion, in which the pulse broadening due to group velocity dispersion is canceled out by the narrowing of the pulse due to chirping due to self-phase modulation caused by the optical force effect. It is. A normal optical pulse undergoes self-phase modulation in proportion to the instantaneous optical power of the optical pulse as described above, so the magnitude of the phase change is different at each point of the pulse, whereas an optical soliton is self-phase modulated over the entire pulse. It has the characteristic that the phase change due to phase modulation is constant.

Therefore, if the clockwise light and counterclockwise light propagating in the optical fiber are set to have a pulse width and optical power that satisfy the conditions for propagating as a soliton, as shown in Fig. 7 (C
), the output light of the boat Pz does not depend on the instantaneous optical power of the incident optical pulse, but rather depends on the energy of the incident optical pulse, so the output energy responds in a step-like manner, and the output waveform is subject to waveform distortion. do not have. Moreover, if a large number of nonlinear Sagnac interferometers satisfying such soliton conditions are connected in series, a steeper dark value characteristic can be obtained.

FIG. 8(a) based on the coupler is a diagram showing an example of the configuration of an optical threshold element based on a nonlinear coupler that can be used in the present invention.
5 is an optical coupler, 21 is an input port, and 22 and 23 are output ports. Input light 13 input to input boat 21
The optical coupler 25 due to its small optical power
When the optical power effect inside can be ignored, it is assumed that all of the light is coupled from the optical waveguide 26 to the optical waveguide 27 and output to the output port 23. When the optical power of the input light 13 becomes stronger,
Due to the optical power effect caused by this, the refractive index of the optical waveguide in the optical coupler 25 changes, and the coupling conditions change, so that a part of the input light is not coupled from the optical waveguide 26 to the optical waveguide 27, and instead is transferred to the boat 22. is output from. Optical power P of light 17 output from boat 22. U, can be expressed by the following formula.

P out= P in [1+cn (π, (P,,
, /PC)'')]/2 (8) Here, P, , l is the input optical power, cn is the Jacobian elliptic function, and PC is the input optical power at which half of the input light is output to the output port 22. , can be expressed by the following equation.

Pe=Aλ/Lcnz (9) Here, n2 is a nonlinear constant, and L is the number of couplers required to couple all input light to adjacent waveguides when the input optical power is small (nonlinear effects can be ignored). length, λ is the wavelength,
A is the effective core cross-sectional area of the waveguide. This situation is shown in FIG. 9, with the horizontal axis representing the optical power of the input light and the vertical axis representing the optical power of the light 17 output from the output port 22. Since the refractive index change due to the optical power effect has a response speed on the femtosecond order, it changes depending on the instantaneous optical power of the incident light of the light 17 output from the output port 22. As is clear from FIG. 9, the optical power of the light 17 outputted from the output port 22 shows a step-like response to the input optical power, so it can be used as an ultra-high-speed dark value element.

In order to make this response steeper, it is necessary to
), a large number of the nonlinear couplers just mentioned can be connected in tandem. The response at this time is shown in FIG. 9, using the number n of columns of the nonlinear coupler as a parameter. It can be seen that the response becomes steeper as the number of columns increases.

The output light from the output port 22 changes depending on the instantaneous optical power of the input light. Since the optical waveform of normal input light is not rectangular, the output light from the output port 22 passes through this optical threshold element and is subjected to waveform distortion at the same time as threshold processing. In order to avoid this, the optical coupler 25 may be configured with an optical waveguide having anomalous dispersion at the wavelength of the input light, and the input light may be propagated as an optical soliton.

(Effects of the Invention) As explained above, the optical discrimination regenerator of the present invention solves the problem of waveform deterioration due to accumulation of noise and jitter and dispersion of optical linear amplifiers, and performs equalization amplification and regeneration in the optical domain. It is possible to provide optical regenerative amplification that performs timing and identification regeneration at ultra-high speed.

[Brief explanation of the drawing]

FIG. 1(a) is a diagram showing the configuration of an embodiment of the present invention.
FIG. 2(b) is a time chart of a light intensity waveform for explaining the present invention in detail; FIG. 2(a) is a diagram showing a configuration example of a stimulated Raman amplifier that can be used as an optical amplifier in the present invention; Figure (b)
is a diagram explaining its operation in the wavelength domain, Figure 3 is a diagram showing the wavelength dependence of the Raman scattering coefficient of glass, and Figure 4 (a) is a diagram of the modulation unstable amplifier that can be used as an optical amplifier in the present invention. A diagram showing a configuration example, FIG. 4(b)
is a diagram explaining its operation in the wavelength domain, Figure 5 is a diagram showing the optical frequency dependence of the modulation unstable gain coefficient as a difference from the angular frequency of the pump light, and Figure 6 is a diagram explaining the optical threshold element in the present invention. FIG. 7 is a diagram showing the optical input/output characteristics of the configuration example shown in FIG. 6. FIG. 8 is a diagram showing an example of the configuration of a nonlinear Sagnac interference type that can be used. FIG. FIG. 9 is a diagram showing a configuration example of a nonlinear coupler, and FIG. 9 is a diagram showing input/output characteristics of the configuration example shown in FIG. 8. 1... Input optical signal pulse train 2... Clock optical pulse 3... Demultiplexed optical clock pulse 4... Regenerated optical signal pulse train 5... Optical multiplexer 6... Optical demultiplexer Vessel 7...
- Optical amplifier 8... Optical threshold element 9... Dichroic mirror 10... Single mode optical fiber 11... Optical filter 12... Amplified optical signal pulse train 13... Optical input 14...・Clockwise light 15... Counterclockwise light 16... Returning light 17
... Output light 18 ... Optical isolator 19 ... Optical coupler 20 ... Optical fiber loop 21 ... Input coupler) 22.23.
... Output boat 24 ... Output light 25 ... Optical coupler 26.27 ... Optical waveguide patent applicant Nippon Telegraph and Telephone Corporation

Claims (1)

[Claims] 1. An optical signal pulse input from an optical transmission line and an optical clock pulse synchronized with the optical signal pulse are input, and only the portion of the optical signal optical pulse where the two overlap in time is amplified. an optical amplifier that outputs amplified light; and an optical threshold element that inputs the optical output of the optical amplifier and outputs light with a nonlinear optical intensity or optical energy with respect to the optical intensity or optical energy of the optical input. An optical identification regenerator characterized by comprising: 2. The optical identification regenerator according to claim 1, wherein the optical amplifier includes a single mode optical fiber, and an optical system that inputs the optical clock pulse and the optical signal pulse into the single mode optical fiber. an optical filter that separates an amplified optical pulse signal from the light emitted from the optical single mode fiber, and the wavelength of the optical pulse signal is determined by the Raman wavelength when the optical clock pulse is input to the single mode optical fiber. An optical identification regenerator characterized by being within a gain band. 3. The optical identification regenerator according to claim 1, wherein the optical amplifier includes a single mode optical fiber having anomalous dispersion at the wavelength of the optical clock pulse, and a single mode optical fiber that transmits the optical clock pulse and the optical signal pulse into the single mode optical fiber. an optical system that enters a mode optical fiber; and an optical filter that separates an amplified optical pulse signal emitted from the optical fiber, and the wavelength of the optical pulse signal is such that the optical clock pulse is An optical discrimination regenerator characterized by being within a modulation instability gain band when input to a fiber. 4. The optical identification regenerator according to any one of claims 1 to 3, wherein the optical threshold element includes an optical Kerr medium in its optical loop, and the optical coupler has a nonlinear sagnac whose branching ratio is not 1:1. (Sagnac) An optical identification regenerator characterized by being composed of interferometers connected in one stage or in multiple stages. 5. In the optical identification regenerator according to any one of claims 1 to 3, the optical threshold element is composed of a nonlinear optical coupler whose branching ratio changes nonlinearly depending on the intensity of the incident light, and the optical threshold element is composed of a nonlinear optical coupler having a nonlinear optical coupler whose branching ratio changes nonlinearly depending on the intensity of the incident light. An optical identification regenerator characterized by having a multi-stage connection configuration.