JP2015161827A - Optical amplification device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a low-noise phase sensitive optical amplifier in which deterioration in a signal SN ratio is smaller than ever before.SOLUTION: An optical amplification device comprises: a first amplifier which amplifies signal light and first fundamental wave light; a first secondary nonlinear optical element which generates a second harmonic from signal light and first fundamental wave light; a multiplexer/demultiplexer which separates only idler light having a specific frequency; a circulator which outputs second fundamental wave light from idler light and laser; a second amplifier and a third amplifier each of which amplifies branched first and second fundamental wave light; a second secondary nonlinear optical element which generates first sum frequency light from first and second fundamental wave light from the second amplifier; a third secondary nonlinear optical element which generates difference frequency light from signal light and first sum frequency light; a fourth secondary nonlinear optical element which generates second sum frequency light from first and second fundamental wave light from the third amplifier; and a fifth secondary nonlinear optical element which performs non-degeneracy parametric amplification from signal light, difference frequency light, and second sum frequency light.

Description

  The present invention relates to an optical amplifying apparatus used in an optical communication system or an optical measurement system.

  In a conventional optical transmission system, an identification regenerative optical repeater that converts an optical signal into an electrical signal and regenerates the optical signal after identifying the digital signal is used to reproduce the signal attenuated by propagating through the optical fiber. It was done. However, this identification / reproduction optical repeater has problems such as limited response speed of electronic components that convert optical signals into electrical signals, and increased power consumption as the speed of transmitted signals increases. .

  As optical amplification means for solving this problem, there are fiber laser amplifiers and semiconductor laser amplifiers that amplify signal light by making excitation light incident on an optical fiber doped with rare earth elements such as erbium and praseodymium. Such fiber laser amplifiers and semiconductor laser amplifiers can amplify signal light as it is, so that there is no limitation on the electrical processing speed that has been a problem in the identification reproduction optical repeater. In addition, the device configuration has the advantage of being relatively simple.

  However, these laser amplifiers do not have a function of shaping a deteriorated signal light waveform. Further, in these laser amplifiers, unavoidably and randomly generated spontaneous emission light is mixed regardless of the signal component, so that the S / N of the signal light is reduced by at least 3 dB before and after amplification. These lead to an increase in transmission code error rate at the time of digital signal transmission, which is a factor of reducing transmission quality.

  As means for overcoming the limitations of the conventional laser amplifier, a phase sensitive amplifier (PSA) has been studied. The PSA has a function of shaping a deteriorated signal light waveform or phase signal due to the influence of dispersion of the transmission fiber. In addition, PSA can suppress spontaneous emission light having a quadrature phase unrelated to the signal and can minimize the spontaneous emission light of the same phase, so that the S / N of the signal light is not deteriorated before and after amplification. In principle it is possible to keep.

  FIG. 1 shows a basic configuration of a conventional PSA. As shown in FIG. 1, the PSA 100 includes a phase sensitive light amplification unit 101 using optical parametric amplification, a pumping light source 102, a pumping light phase control unit 103, and first and second light branching units 104-1. And 104-2. As shown in FIG. 1, the signal light 110 input to the PSA 100 is branched into two by the optical branching unit 104-1, and one enters the phase sensitive light amplification unit 101 and the other enters the excitation light source 102. . The phase of the excitation light 111 emitted from the excitation light source 102 is adjusted via the excitation light phase control unit 103 and enters the phase sensitive light amplification unit 101. The phase sensitive amplification unit 101 outputs an output signal light 112 based on the input signal light 110 and the excitation light 111.

  The phase sensitive light amplifying unit 101 amplifies the signal light 110 when the phase of the incident signal light 110 and the phase of the excitation light 111 coincide with each other. It has a characteristic to attenuate. If the phase between the pumping light 111 and the signal light 110 is matched so that the amplification gain is maximized using this characteristic, the spontaneous emission light having the quadrature phase with the signal light 110 is not generated, and the in-phase component is also generated. Excessive spontaneous emission light beyond the noise of signal light is not generated. Therefore, the signal light 110 can be amplified without degrading the S / N ratio.

  In order to achieve such phase synchronization between the signal light 110 and the pump light 111, the pump light phase control unit 103 is synchronized with the phase of the signal light 110 branched by the first light branching unit 104-1. The phase of the excitation light 111 is controlled. In addition, the pumping light phase control unit 103 detects a part of the output signal light 112 branched by the second optical branching unit 104-2 with a narrow-band detector, and the amplification gain of the output signal light 112 is maximized. Thus, the phase of the excitation light 111 is controlled. As a result, the phase sensitive optical amplifying unit 102 realizes optical amplification without degradation of the S / N ratio based on the above principle.

  The pumping light phase control unit 103 may be configured to directly control the phase of the pumping light source 102 in addition to the configuration of controlling the phase of the pumping light 111 on the output side of the pumping light source 102. Further, when the light source that generates the signal light 110 is disposed near the phase-sensitive light amplification unit 101, a part of the signal light source can be branched and used as excitation light.

T. Umeki et al., “Phase sensitive degenerate parametric amplification using directly-bonded PPLN ridge waveguides,” Optics Express, 2011, Vol. 19, No. 7, pp. 6326-6332 J. Kakande et al., “First demonstration of all-optical QPSK signal regeneration in a novel multi-format phase sensitive amplifier,” Post Deadline paper, ECOC 2010, 2010 M. Asobe et al., “In-line phase-sensitive amplifier for QPSK signal using multiple QPM LiNbO3 waveguide,” In Proceedings of the OptoElectronics Communications Conference, Post Deadline paper, 2010, PD2-3

However, the above-described prior art has the following problems. As the nonlinear optical medium for performing the parametric amplification described above, a method using a second-order nonlinear optical material typified by a periodically poled LiNbO ₃ (PPLN) waveguide and a method using a third-order nonlinear optical material typified by a silica glass fiber There is.

  FIG. 2 illustrates a configuration of a conventional PSA using a PPLN waveguide disclosed in Non-Patent Document 1 and the like. 2 includes an erbium-doped fiber laser amplifier (EDFA) 201, first and second second-order nonlinear optical elements 202 and 204, and first and second optical branching sections 203-1 and 203-. 2, a phase modulator 205, an optical fiber stretcher 206 using PZT, a polarization maintaining fiber 207, a photodetector 208, and a phase locked loop (PLL) circuit 209. The first second-order nonlinear optical element 202 includes a first spatial optical system 211, a first PPLN waveguide 212, a second spatial optical system 213, and a first dichroic mirror 214. The second-order nonlinear optical element 204 includes a third spatial optical system 215, a second PPLN waveguide 216, a fourth spatial optical system 217, a second dichroic mirror 218, and a third dichroic mirror 219. .

  The first spatial optical system 211 couples light input from the input port of the first second-order nonlinear element 202 to the first PPLN waveguide 212. The second spatial optical system 213 couples the light output from the first PPLN waveguide 212 to the output port of the first second-order nonlinear optical element 202 via the first dichroic mirror 214. The third spatial optical system 215 couples light input from the input port of the second second-order nonlinear optical element 204 to the second PPLN waveguide 216 via the second dichroic mirror 218. The fourth spatial optical system 217 couples the light output from the second PPLN waveguide 216 to the output port of the second second-order nonlinear optical element 204 via the third dichroic mirror 219.

  In the example shown in FIG. 2, the signal light 250 incident on the PSA 200 is branched by the optical branching unit 203-1, one incident on the second second-order nonlinear optical element 204, and the other as the excitation fundamental wave light 251. The phase is controlled via the phase modulator 205 and the optical fiber expander 206 and enters the EDFA 201. In order to obtain sufficient power to obtain the nonlinear optical effect from the weak laser beam used for optical communication, the EDFA 201 amplifies the incident excitation fundamental wave light 251, and the amplified excitation fundamental wave light 251 is converted into the first secondary light. The light enters the nonlinear optical element 202-1. In the first second-order nonlinear optical element 202-1, a second harmonic (hereinafter referred to as “SH light”) 252 is generated from the incident excitation fundamental wave light 251, and the generated SH light 252 passes through the polarization maintaining fiber 207. The light enters the second second-order nonlinear optical element 204. The second second-order nonlinear optical element 204 performs phase-sensitive amplification by performing degenerate parametric amplification with the incident signal light 250 and the SH light 252, and outputs the output signal light 253.

In PSA, in order to amplify only light in phase with a signal, it is necessary that the phase of signal light and pumping light coincide with each other or be shifted by π radians as described above. That is, when the second-order nonlinear optical effect is used, it is necessary that the phase φ _2ωs of the excitation light and the phase φ _{ωs of the} signal light, which has a wavelength corresponding to the SH light, satisfy the following relationship (Equation 1). Become.
_{Δφ = 1/2 (φ 2ωs} -φ ωs) = nπ ( where, n is an integer) (Equation 1)

  FIG. 3 is a graph showing the relationship between the phase difference Δφ between the input signal light and the pump light and the gain (dB) in the conventional PSA using the second-order nonlinear optical effect. It can be seen that the gain is maximum when Δφ is −π, 0, or π.

  In the configuration shown in FIG. 2, in order to phase-synchronize the signal light 250 and the excitation fundamental wave light 251, the phase modulator 205 is used to perform phase modulation on the excitation fundamental wave light 251 with a weak pilot signal, and then output. A part of the signal light 253 is branched and detected by the detector 208. Since this pilot signal component is minimized when the phase difference Δφ shown in FIG. 3 is minimized, the phase locked loop circuit ( PLL) 209 is used to provide feedback to the optical fiber stretcher 206. The phase of the excitation fundamental light 251 can be controlled to achieve phase synchronization between the signal light 250 and the excitation fundamental light 251.

  In the configuration in which the PPLN waveguide is used as a nonlinear medium and the signal light 250 and the SH light 252 are incident on the second-order nonlinear optical element 204 to perform degenerate parametric amplification, the SH light 252 is generated once. When performing parametric amplification, for example, the components of the excitation fundamental light are removed using the characteristics of the dichroic mirrors 206-1 and 206-2, so that only the SH light 252 and the signal light 250 are converted into the second second-order nonlinear optical element 204. It can inject into a parametric amplification medium like. Therefore, noise due to mixing of spontaneous emission light generated by the EDFA 201 can be prevented, so that low-noise optical amplification can be performed.

  As described above, the PPLN waveguide is used as a nonlinear optical medium, and the nonlinear medium is excited using the SH light 252 so that low-noise phase-sensitive amplification can be achieved without being affected by noise generated by the EDFA 201. In addition, the phase noise can be reduced by utilizing the characteristic of attenuating the quadrature component.

  However, as shown in FIG. 3, the conventional configuration method has a characteristic of attenuating orthogonal phase components, so that an ordinary intensity-modulated signal or binary phase modulation such as IMDD, BPSK, or DPSK is used. Although it can be used to amplify the modulation signal, it is not possible to amplify signals such as QPSK (four values) and 8PSK which are multi-level modulation formats.

  On the other hand, as disclosed in Non-Patent Documents 2 and 3 and the like, it is known that a phase regenerative amplification can be performed by phase-sensitive amplification of a signal such as QPSK. Non-Patent Documents 2 and 3 each show a method using a quartz glass fiber which is a third-order nonlinear optical material and a method using PPLN which is a second-order nonlinear optical material. FIG. 4 illustrates a PSA using a quartz glass fiber that is performing QPSK phase reproduction amplification, which is four-level phase modulation.

  In FIG. 4, the first and second EDFAs 401 and 408, the first excitation light source 402, the first to fifth optical multiplexing / demultiplexing cells 403-1 to 403-5, and the first optical fiber 404 are illustrated. , A demultiplexer 405, an optical circulator 406, a second pumping light source 407 made of a semiconductor laser, a second optical fiber 409, a bandpass filter 410, a PLL circuit 412, and an optical fiber stretcher 413 using PZT. A PSA 400 with is shown.

  In the example shown in FIG. 4, the signal light 415 input to the PSA 400 is amplified by the first EDFA 401 and then enters the first optical multiplexer / demultiplexer 403-1. The first excitation light 414 generated by the first excitation light source 402 is branched by the second optical multiplexer / demultiplexer 403-2, one of which is incident on the first optical multiplexer / demultiplexer 403-1 and the other is The light enters the third optical multiplexer / demultiplexer 403-3. A part of the first pumping light 414 incident on the first optical multiplexer / demultiplexer 403-1 and the amplified signal are multiplexed by the first optical multiplexer / demultiplexer 403-1, The light enters the optical fiber 404.

  FIG. 5A shows the frequency arrangement in the first optical fiber 404. When the signal combined by the first optical multiplexer / demultiplexer 403-1 enters the first optical fiber 404, as shown in FIG. 5A, four-wave mixing in the first optical fiber 404 is performed. (Hereinafter, FWM) generates a plurality of signal groups. This process is described in detail below.

  In the first optical fiber 404, first the first pumping light 414 is converted into the first idler light 416 by degenerate FWM using the signal light 415 as pumping light. Next, the signal light 415 is converted into the second idler light 417 by degenerate FWM using the first idler light 416 as excitation light. Next, the first idler light 416 is converted into the third idler light 418 by degenerate FWM using the second idler light 417 as excitation light.

  As described above, the plurality of idler lights 416 to 418 are sequentially generated in the first optical fiber 404 by the secondary FWM because the zero dispersion wavelength of the first optical fiber 404 is distributed in the length direction. It depends. That is, in the degenerate FWM, normally, the wavelength of the first optical fiber 404 becomes the zero dispersion wavelength at the wavelength that serves as the pumping light, so that the phase matching of the FWM process is achieved and wavelength conversion is performed.

However, since the structure of the optical fiber is not completely uniform in the length direction, the zero dispersion wavelength generally changes in the length direction. For this reason, as described above, the wavelengths of the plurality of idler lights 416 to 418 become wavelengths that act as excitation light, respectively, and cause secondary FWM. At this time, the phase of idler light 416 to 418 generated in each process is given by the following (Expression 2) to (Expression 4) from the phase matching condition in each FWM process.
φ _i1 = 2φ _s −φ _p1 (Formula 2)
φ _i2 = 2φ _i1 −φ _s = 3φ _s −2φ _p1 (Formula 3)
φ _i3 = 2φ _i2 −φ _i1 = 4φ _s -3φ _p1 (Formula 4)

Here, in the above (Expression 2) to (Expression 4), φ _s is the phase of the signal light 415, φ _p1 is the phase of the first excitation light 414, φ _i1 is the phase of the first idler light 416, φ _i2 Represents the phase of the second idler light 417, and φ _i3 represents the phase of the third idler light 418.

  If attention is paid to (Equation 4) assuming that the phase of the first excitation light 414 is constant, it can be seen that the phase of the third idler light 418 is four times that of the signal light 415. Therefore, it can be understood that the phase of the third idler light 418 is constant and the phase of the carrier wave can be extracted from the QPSK signal when the signal has only an integer multiple of π / 2, such as QPSK.

  Returning to FIG. Each signal generated by the first optical fiber 404 is routed for each wavelength by the demultiplexer 405, and the third idler light 418 enters the optical circulator 406, and the second excitation is performed in the optical circulator 406. It is injection-synchronized with the laser output from the light source 407, and the remaining intensity modulation component is removed and output as the second excitation light 419.

The first pumping light 414 and the second pumping light 419 are multiplexed by the third optical multiplexer / demultiplexer 403-3, amplified by the EDFA 408, and then the signal light 415 and the second optical beam output from the demultiplexer 405. The second idler light 417 and the fourth optical multiplexer / demultiplexer 403-4 are combined and incident on the second optical fiber 409. In the second optical fiber 409, phase sensitive amplification by FWM is performed. At this time, when attention is paid to the phases of the four lights incident on the second optical fiber 409, it is understood that the following (formula 5) and (formula 6) are established.
φ _p2 = φ _i3 = 4φ _s -3φ _p1 (Formula 5)
φ _p1 + φ _p2 = φ _s + φ _i2 = 4φ _s -2φ _p1 (Formula 6)
Here, in (Equation 6), φ _p2 is the phase of the second excitation light 419. As can be seen from (Expression 6), the sum of the phases of the first pump light 414 and the second pump light 419 and the sum of the phases of the signal light 415 and the second idler light 417 coincide. Accordingly, the phase matching condition is satisfied among the four input lights in the second optical fiber 409.

  FIG. 5B shows the frequency arrangement in the second optical fiber 409. As shown in FIG. 5B, the energy of the first excitation light 414 and the second excitation light 415 is converted into the signal light 415 and the second idler light 417, and optical parametric amplification is performed. At this time, since (Equation 6) holds only when the phase of the signal is an integral multiple of π / 2, when a QPSK signal is incident, only signals in four phase states are phase-sensitive amplified. Phase recovery amplification of the QPSK signal is achieved.

  In the actual amplification operation, since the optical phase fluctuates due to the expansion and contraction of the optical fiber components, only the signal light amplified from the output of the second optical fiber 409 is taken out by the band pass filter 410 and a part thereof is taken as the fifth. Are detected by the optical multiplexer / demultiplexer 403-5 and the optical detector 411, and fed back to the optical fiber expander 413 made of PZT via the PLL circuit 412, thereby stabilizing the phase between the signal and the pumping light, and the phase sensitivity. Amplification is achieved.

  However, in the configuration using the above optical fiber as a nonlinear optical medium, as can be seen from the frequency arrangement in FIGS. 5A and 5B, strong excitation light in the same wavelength region as the weak signal. In addition, since there are two of them, and their pumping light is amplified by the optical fiber amplifier, it is inevitable that the spontaneous emission light generated by the optical fiber amplifier is mixed into the signal wavelength. Furthermore, in order to obtain a sufficient gain, the length of the optical fiber is as long as several hundreds of meters, which is not practical. Further, when strong CW excitation light is incident on the optical fiber, it is caused by stimulated Brillouin scattering. Since light power above a certain level cannot be incident due to backscattering, the threshold of stimulated Brillouin scattering is lowered by intentionally adding phase modulation to increase the line width of the pumping light or by distributing the tension applied to the optical fiber. Therefore, there is a problem that a new noise is generated due to such an extra mechanism and the configuration is complicated.

  In view of the above-described problems of the prior art, Non-Patent Document 3 discloses a phase regenerative optical amplifier using a PPLN waveguide for a QPSK signal that is smaller, lower noise, and simpler than conventional ones. Yes. FIG. 6 illustrates a PSA using a PPLN waveguide performing QPSK phase reproduction amplification, which is quaternary phase modulation.

  6 includes first and second EDFAs 601 and 608, first and second excitation light sources 602 and 607, first to fourth optical multiplexers / demultiplexers 603-1 to 603-4, To third-order nonlinear optical elements 604, 610, and 611, a demultiplexer 605, first and second bandpass filters 609 and 612, a photodetector 613, a PLL circuit 614, and light generated by PZT. An optical amplification device 600 comprising a fiber stretcher 615 and a polarization maintaining optical fiber 616 is shown. The first second-order nonlinear optical element 604 includes a first spatial optical system 621, a first PPLN waveguide 622, a second spatial optical system 623, and a first dichroic mirror 624. The second second-order nonlinear optical element 610 includes a third spatial optical system 625, a second PPLN waveguide 626, a fourth spatial optical system 627, and a second dichroic mirror 628. The third second-order nonlinear optical element 611 includes a fifth spatial optical system 629, a third PPLN waveguide 630, a sixth spatial optical system 631, a third dichroic mirror 632, and a fourth dichroic mirror. 633.

  The first spatial optical system 621 couples light input from the input port of the first second-order nonlinear optical element 604 to the first PPLN waveguide 622. The second spatial optical system 623 couples the light output from the first PPLN waveguide 622 to the output port of the first second-order nonlinear optical element 604 via the first dichroic mirror 624. The third spatial optical system 625 couples light input from the input port of the second second-order nonlinear optical element 610 to the second PPLN waveguide 626. The fourth spatial optical system 627 couples the light output from the second PPLN waveguide 626 to the output port of the second second-order nonlinear optical element 610 via the second dichroic mirror 628. The fifth spatial optical system 629 couples light input from the input port of the third second-order nonlinear optical element 611 to the third PPLN waveguide 630 via the third dichroic mirror 632. The sixth spatial optical system 631 couples the light output from the third PPLN waveguide 630 to the output port of the third second-order nonlinear optical element 611 via the fourth dichroic mirror 632.

  In the example shown in FIG. 6, the signal light 652 that has been QPSK modulated in the 1.55 μm band and has entered the optical amplifying apparatus 600 is incident on the first optical multiplexer / demultiplexer 603-1. The first pumping light 651 generated by the first pumping light source 602 is branched by the second optical multiplexer / demultiplexer 603-2, one of which is incident on the first optical multiplexer / demultiplexer 603-1, and the other is The light enters the third optical multiplexer / demultiplexer 603-3. A part of the first pumping light 651 and the signal light 652 incident on the first optical multiplexer / demultiplexer 603-1 are multiplexed by the first optical multiplexer / demultiplexer 603-1, and then input to the first EDFA 601. The light is incident and amplified, and is incident on the first second-order nonlinear optical element 604.

  Here, in order to extract a carrier wave from a QPSK signal in which the carrier wave disappears by using a compact PPLN waveguide, the first second-order nonlinear optical element 604 includes three pseudo-wires at a frequency interval of 100 GHz in a 1.55 μm band. A first PPLN waveguide 622 having a phase matching wavelength is fabricated.

Here, the wavelength conversion process in the first PPLN waveguide 622 will be described. FIG. 7A shows the frequency arrangement and phase matching characteristics in the first PPLN waveguide 622. As shown in FIG. 7A, the lowest frequency side peak among the three phase matching peaks of the first PPLN waveguide 622 is matched with the frequency of the signal light 652. The wavelength of the first excitation light 651 is set so as to be detuned to a frequency lower by 100 GHz than the signal light 652. Under the above conditions, in the first PPLN waveguide 622, the signal light 652 is first converted into SH light, and the first excitation light 651 is generated by difference frequency generation (DFG) using the SH light as excitation light. The first idler light 653 is converted. Since the frequency of the converted first idler light 653 matches the second phase matching wavelength of the first PPLN waveguide 622, the first idler light 653 is converted into SH light, and the SH light is converted into excitation light. As a result, the signal light 652 is converted into the second idler light 654. Since the frequency of the second idler light 654 matches the third phase matching wavelength of the first PPLN waveguide 622, the second idler light 654 is converted into SH light, and the SH light is used as the excitation light. One idler light 653 is converted into a third idler light 655. At this time, the phases of the first to third idler lights 653 to 655 are given by the following equations.
φ _i1 = 2φ _s −φ _p1 (Formula 7)
φ _i2 = 2φ _i1 −φ _s = 3φ _s −2φ _p1 (Equation 8)
φ _i3 = 2φ _i2 −φ _i1 = 4φ _s -3φ _p1 (Equation 9)

Here, in the above (Expression 7) to (Expression 9), φ _s is the phase of the signal light 652, φ _φ1 is the phase of the first excitation light 651, φ _i1 is the phase of the first idler light 653, φ _i2 Represents the phase of the second idler light 654, and φ _i3 represents the phase of the third idler light 655.

  Now, if it is assumed that the phase of the first excitation light 651 is constant, paying attention to (Equation 9), it can be seen that the phase of the third idler light 655 is four times the phase of the signal light 652. Therefore, it can be understood that the phase of the third idler light 655 is constant in a signal having only an integer multiple of π / 2, such as QPSK, and the phase of the carrier wave can be extracted from the QPSK signal.

  Referring to FIG. 6, the output of the first second-order nonlinear optical element 604 having the first PPLN waveguide 622 is separated into two by the wavelength demultiplexer 605, and the signal light 652 and the second light are separated. The idler light 654 enters the third second-order nonlinear optical element 611, and the third idler light 655 enters the optical circulator 606. The third idler light 655 is injection-locked to the laser output from the second excitation light source 607 (semiconductor laser) in the optical circulator 606, and the remaining intensity modulation component is removed and output as the second excitation light 656. . The first pumping light 651 and the second pumping light 656 are combined by the third optical multiplexer / demultiplexer 603-3, amplified by the second EDFA 608, and then the first pumping light 651 and the second pumping light 656. After excessive spontaneous emission light is removed by a bandpass filter 609 that transmits only the light 656, the light is incident on the second PPLN waveguide 626 of the second second-order nonlinear optical element 610.

The second PPLN waveguide 626 is manufactured by a method similar to that of the first PPLN waveguide 622, but the second PPLN waveguide 626 is not subjected to phase modulation in a domain-inverted structure having a period of 17 μm. Therefore, the phase matching wavelength when evaluated by second harmonic generation (SHG) is the frequency arrangement in the second PPLN waveguide 626 of the second second-order nonlinear optical element 610 shown in FIG. Thus, it matches the wavelength of the first idler light 653. Both the first pumping light 651 and the second pumping light 656 are separated from each other in the opposite direction by 200 GHz from the phase matching wavelength. Therefore, the phase matching condition for the sum frequency generation of the first pumping light 651 and the second pumping light 656 is satisfied, and the second PPLN waveguide 626 corresponds to the SH light of the first idler light 653. Wavelength conversion is performed on the wavelength as sum frequency light 657. At this time, the phase φ _SF of the sum frequency light 657 is given by the following (formula 10). Here, φ _p2 indicates the phase of the second excitation light 656.
φ _SF = φ _p1 + φ _p2 = −2φ _p1 (Formula 10)

  The sum frequency light 657 is separated from the first pumping light 651 and the second pumping light 656 by the second dichroic mirror 628 built in the second second-order nonlinear optical element 610, and then, the sum frequency light 657. The light enters the third PPLN waveguide 630 of the third second-order nonlinear optical element 611 through the polarization maintaining optical fiber 616 in the 78 μm band.

  On the other hand, the signal light 652 and the second idler light 654 separated by the branching filter 605 and incident on the third second-order nonlinear optical element 611 are also transmitted via the third dichroic mirror 632 to the third PPLN waveguide. 630 is incident. The third PPLN waveguide 630 has characteristics equivalent to those of the second PPLN waveguide 626, and therefore the phase matching wavelength thereof is substantially the same as that of the second PPLN waveguide 626, and the phase matching wavelengths of both are It is set to match by appropriate temperature adjustment.

FIG. 7 (c) shows the frequency arrangement in the third PPLN waveguide 630. As shown in FIG. 7C, the frequency of the sum frequency light 657 corresponds to the SH light having the phase matching wavelength of the third PPLN waveguide 630 in the 1.55 μm band. Therefore, the phase matching condition is satisfied between the sum frequency light 657, the signal light 652, and the second idler light 654, and when the sum of the phase of the signal light 652 and the phase of the second idler light 654 is calculated, It is expressed as (Equation 11) below.
φ _s + φ _i2 = 4φ _s -2φ _p1 (Formula 11)

  Comparing (Equation 11) with (Equation 10), it can be seen that the two match only when the phase of the signal is an integral multiple of π / 2, so that the QPSK signal can be phase-sensitively amplified.

  Unlike PSA using conventional optical fibers, there is no strong excitation light in the 1.55 μm band, and there is no spontaneous emission light generated by the EDFA used for excitation light. Phase sensitive amplification can be performed while maintaining.

  In the actual amplification operation, only the signal light amplified from the output of the third PPLN waveguide 630 is used to suppress the influence of the phase fluctuation due to the fluctuation of the optical path length due to the expansion and contraction of the optical fiber connecting each optical component. Is extracted using a bandpass filter 612, and a part of the signal is branched using a fourth optical multiplexer / demultiplexer 603-4, detected by a photodetector 613, and then phase-modulated via a PLL circuit 614. -Feedback is provided to the optical fiber stretcher 615 made of PZT to achieve stable operation.

  However, in the configuration of the conventional PSA for the signal light having the modulation format multi-valued in the phase direction described above, the optical amplifying medium from the input regardless of whether the third-order nonlinear optical element or the second-order nonlinear optical element is used. (The second optical fiber 409 in the configuration example of the third-order nonlinear optical element shown in FIG. 4 and the third PPLN waveguide 630 in the case of the second-order nonlinear optical element shown in FIG. 6). -There was a problem that the noise ratio (SN ratio) deteriorated and included factors that deteriorated the characteristics of the entire system. The main factor is that the second idler light and the third idler light generated using one nonlinear optical medium must be used.

  In order to realize the PSA for the QPSK signal, the third idler light is used to cancel the phase modulation component and regenerate the carrier wave phase, and the second idler light is used as a pair of signal light that interacts with the signal light. Used. The second idler light and the third idler light are generated from the signal light and the first excitation light through a multistage nonlinear process using one nonlinear optical medium. However, since the power of signal light used in normal optical communication is weak, the conversion efficiency decreases as the number of nonlinear processes increases. That is, normally, the first idler light is less than the signal light intensity, the second idler light is less intense than the first idler light, and the third idler light is stronger than the second idler light. It gets smaller.

  In order to realize the PSA for the QPSK signal, it is necessary to input the signal light and the second idler light together with the excitation light to the optical amplifier. In order to realize ideal phase-sensitive amplification, the light intensity of the signal light and the second idler light needs to be equal. For this reason, when the light intensity of the generated second idler light is weaker than the signal light, it is necessary to attenuate the signal light alone and equalize the light intensity with the second idler light. However, generally, after the signal light is once attenuated before the input of the optical amplifier (here, the second optical fiber 409 in the configuration example of FIG. 4 and the third PPLN waveguide 630 in the configuration example of FIG. 6). Since the optical amplifier has a characteristic that when amplified, the SN ratio deteriorates by the amount of attenuation, the equalization of the light intensity with the second idler light immediately takes the SN ratio of the signal light as considered in the entire optical amplifier. It will be deteriorated.

  One method for increasing the light intensity of the first to third idler lights is to increase the intensity of the signal light and the first excitation light input to the nonlinear optical medium. For this reason, in the conventional configuration, at least the signal light is amplified by the EDFA and then enters the nonlinear element (the first optical fiber 404 in the configuration example of FIG. 4 and the first PPLN waveguide 622 in the configuration example of FIG. 6). . However, in laser amplifiers represented by EDFA, spontaneously emitted light inevitably and randomly generated is mixed regardless of the signal component, so the S / N of the signal light called the standard quantum limit of the optical amplifier is around before and after amplification. There was a problem that at least 3 dB lower. For this reason, if the signal light is amplified by an EDFA or the like, at that time, at least the S / N deteriorates by 3 dB, and it becomes difficult to realize a low-noise amplifier.

  Further, since the second idler light and the third idler light are generated by using one nonlinear optical medium, a demultiplexer that separates these lights after generation (the demultiplexer in the configuration example of FIG. 4). 405 and the configuration example of FIG. 6 require a duplexer 605). The duplexer outputs the signal light and the second idler light to one port, outputs only the third idler light to the other port, and quenches the first idler light and the first excitation light. It is necessary to have such a complicated function. The use of such a high-function duplexer has a problem that the cost of the entire amplifier is significantly increased and the reliability is lowered. Further, since complicated demultiplexing is performed, there is a problem in that the optical loss of the demultiplexer becomes large. Since the optical loss of the demultiplexer is before the input to the optical amplifying medium, as described above, the optical loss of the demultiplexer immediately causes the S / N of the entire system to deteriorate.

  As described above, in the conventional configuration in which idler light is sequentially generated using one nonlinear optical medium, excess noise is mixed by the laser amplifier for increasing the generation efficiency, and the light intensity of the signal light and the second idler light is increased. There have been problems such as S / N degradation due to equalization of light and optical loss due to a duplexer having a complex demultiplexing function.

In order to solve the above-mentioned problem, an optical amplifying device according to claim 1 of the present invention is a multi-value phase that takes M (M is an integer of 4 or more) types of values by optical mixing using a second-order nonlinear optical effect. A phase sensitive optical amplifying device for amplifying modulated signal light, wherein a first optical multiplexer / demultiplexer for branching the signal light having a frequency ω _s and a first fundamental light having a frequency ω _LO1 are generated. The first light source, the second optical multiplexer / demultiplexer for branching the first fundamental light, the signal light output from one output terminal of the first optical multiplexer / demultiplexer, and the second light A first optical fiber laser amplifier that inputs, amplifies and outputs the first fundamental wave light output from one of the optical multiplexer / demultiplexers, and satisfies the quasi-phase matching at a plurality of wavelengths, and is amplified A signal light and a first fundamental wave light are input, and a second harmonic is generated and output. The second-order nonlinear optical element, and the multiplexer / demultiplexer that separates and outputs only idler light having a frequency of Mω _s − (M−1) ω _LO1 from the output of the first second-order nonlinear optical element, A second light source that outputs a laser that can be injection-locked to the idler light output from the duplexer, and the laser that outputs the idler light output from the duplexer from the second excitation light source. An optical circulator that outputs the second fundamental wave light in synchronization with the first optical wave, the first fundamental wave light output from the other output terminal of the second optical multiplexer / demultiplexer, and the optical circulator output from the optical circulator. A third optical multiplexer / demultiplexer for inputting and multiplexing the second fundamental wave light, branching and outputting the first fundamental wave light and the second fundamental wave light, and the third optical multiplexer; The first basic output from one output terminal of the duplexer A second optical fiber laser amplifier that amplifies and outputs the light and the second fundamental wave light, and the first fundamental wave light and the second fundamental wave light that are output from the second optical fiber laser amplifier. A second second-order nonlinear optical element that generates and outputs the sum frequency light, the signal light output from the other output terminal of the first optical multiplexer / demultiplexer, and the first sum frequency light. A third-order nonlinear optical element that generates a difference frequency light by inputting and outputs the signal light and the difference frequency light, and the first output from the other output terminal of the third optical multiplexer / demultiplexer. A third optical fiber laser amplifier that amplifies and outputs the first fundamental wave light and the second fundamental wave light, and the first fundamental wave light and the second fundamental wave light output from the third optical fiber laser amplifier. 4th secondary nonlinearity that generates and outputs second sum frequency light from A fifth-order nonlinear optical element that performs non-degenerate parametric amplification by inputting the shape optical element, the signal light, the difference frequency light, and the second sum frequency light. To do.

  An optical amplifying device according to a second aspect of the present invention is the optical amplifying device according to the first aspect, wherein the second optical fiber laser amplifier and the second second-order nonlinear optical element are A first bandpass filter that transmits only the first fundamental wave light and the second fundamental wave light is provided.

  An optical amplifying device according to a third aspect of the present invention is the optical amplifying device according to the first or second aspect, wherein the second secondary is supplied from the one output terminal of the first optical multiplexer / demultiplexer. An optical path from the other output terminal of the first optical multiplexer / demultiplexer to the third second-order nonlinear optical element via a nonlinear optical element to the third second-order nonlinear optical element A first delay controller for equalizing the length is further provided.

An optical amplification device according to a fourth aspect of the present invention is the optical amplification device according to any one of the first to third aspects, wherein the first second-order nonlinear optical element has a periodically poled structure. _The periodic polarization inversion structure is subjected to spatial phase modulation or period modulation with a period longer than the polarization inversion period, and the first second-order nonlinear optical element is at least The quasi phase matching wavelength matches the frequency of ω _s , 2ω _s −ω _LO1 , and the QPSK signal is amplified.

  An optical amplification device according to claim 5 of the present invention is the optical amplification device according to any one of claims 1 to 3, wherein the signal light is branched and the signal light is supplied from one output terminal. The fourth optical multiplexer / demultiplexer output to the first optical multiplexer / demultiplexer and the signal light output from the other output terminal of the fourth optical multiplexer / demultiplexer are input to delay the signal light A second delay controller; a second bandpass filter that transmits only the difference frequency light out of the signal light and the difference frequency light output from the third second-order nonlinear optical element; 2 to combine the signal light output from the delay control circuit 2 and the difference frequency light output from the second bandpass filter, and output the resultant to the fifth second-order nonlinear optical element. And a waver.

  An optical amplification device according to a sixth aspect of the present invention is the optical amplification device according to the fifth aspect, wherein the second delay controller is the one output terminal of the fourth optical multiplexer / demultiplexer. Through the second second-order nonlinear optical element and the third second-order nonlinear optical element to the wavelength multiplexer, and the wavelength from the other output terminal of the fourth optical multiplexer / demultiplexer The optical path to the multiplexer is configured to have the same length.

An optical amplification apparatus according to a seventh aspect of the present invention is the optical amplification apparatus according to the fifth or sixth aspect, wherein the first second-order nonlinear optical element is made of LiNbO ₃ having a periodically poled structure. The periodic polarization reversal structure is subjected to spatial phase modulation or periodic modulation with a period longer than the polarization reversal period, and the first second-order nonlinear optical element includes at least ω _s , The quasi phase matching wavelength matches the frequency of 2ω _s −ω _LO1 , 4ω _s −3ω _LO1 , and the 8PSK signal is amplified.

  According to the optical amplifier of the present invention, the pumping light is once converted into the second harmonic wavelength and then combined with the signal light to perform parametric amplification, so that the pumping light can be easily separated. Since idler light used for interaction can be generated with high efficiency and degradation of the SN ratio can be suppressed, low-noise amplification becomes possible. Furthermore, according to the optical amplifier according to the present invention, since the signal light is not amplified, it is possible to suppress degradation of the SN ratio due to mixing of spontaneous emission light generated by the optical fiber amplifier used internally. Further, according to the optical amplifier according to the present invention, the duplexer according to the present invention only needs to have a function of separating only the third idler light. A decline in sex can be avoided.

  As a result, the optical amplifier according to the present invention can perform optical amplification while suppressing the deterioration of the SN ratio and can perform amplification while reproducing phase information. Therefore, relay amplification can be performed while reducing phase noise.

It is explanatory drawing of a structure of the conventional phase sensitive optical amplifier. It is explanatory drawing of a structure of the phase sensitive optical amplifier using the conventional secondary nonlinear optical effect. It is a graph which shows the relationship between phase difference (DELTA) (phi) between input signal light-excitation light, and a gain in the phase sensitive optical amplifier using the conventional secondary nonlinear optical effect. It is explanatory drawing of a structure of the phase sensitive optical amplifier using the conventional 3rd order nonlinear optical effect. It is a figure which shows the frequency arrangement | positioning of the light in the optical fiber of the conventional PSA. It is explanatory drawing of a structure of the phase sensitive optical amplifier using the conventional secondary nonlinear optical effect. It is a figure which shows the frequency arrangement | positioning and phase matching characteristic of the light in each secondary nonlinear optical element of the conventional PSA. It is a figure for demonstrating the structure of the optical amplifier which concerns on Example 1 of this invention. It is a figure for demonstrating operation | movement of the optical amplifier which concerns on Example 1 of this invention. It is a figure which shows the optical spectrum of the 4th idler light produced | generated in the 3rd second-order nonlinear optical element of the optical amplifier which concerns on Example 1 of this invention. It is a figure which shows the optical spectrum of the signal light and the 4th idler light which were amplified by the optical amplifier which concerns on Example 1 of this invention. It is a constellation map in the input / output of the signal amplified by the optical amplifier which concerns on Example 1 of this invention. It is a figure for demonstrating the structure of the optical amplifier which concerns on Example 2 of this invention. It is a constellation map in the input-output of the signal amplified by the optical amplifier which concerns on Example 2 of this invention.

  Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the drawings.

Example 1
In the optical amplifying device according to the first embodiment, a carrier wave is extracted from the QPSK signal, and after generating excitation light phase-synchronized with the carrier wave, idler light necessary for interaction with the signal light is generated with high efficiency. A PSA using a PPLN waveguide was constructed.

  FIG. 8 shows the configuration of the optical amplifying device according to Embodiment 1 of the present invention. 8 includes first to third EDFAs 801, 808, and 813, first and second fundamental wave light sources 802 and 807, first to fifth optical multiplexers 803-1 to 803-5, First to fifth second-order nonlinear optical elements 804, 810, 811, 814, 815, a demultiplexer 805, an optical circulator 806, first to third bandpass filters 809, 813, 816, and light An optical amplifying apparatus 800 including a detector 817, a PLL circuit 818, an optical fiber stretcher 819 based on PZT, and a delay controller 820 is shown.

  The first second-order nonlinear optical element 804 includes first and second spatial optical systems 821 and 823, a first PPLN waveguide 820, and a first dichroic mirror 824. The second second-order nonlinear optical element 810 includes third and fourth spatial optical systems 825 and 827, a second PPLN waveguide 826, and a second dichroic mirror 828. The third second-order nonlinear optical element 811 includes fifth and sixth spatial optical systems 829 and 831, a third PPLN waveguide 830, and third and fourth dichroic mirrors 823 and 833. The fourth second-order nonlinear optical element 814 includes seventh and eighth spatial optical systems 834 and 836, a fourth PPLN waveguide 835, and a fifth dichroic mirror 837. The fifth second-order nonlinear optical element 815 includes ninth and tenth spatial optical systems 838 and 840, a fifth PPLN waveguide 839, and sixth and seventh dichroic mirrors 841 and 842.

  The first spatial optical system 821 couples light input from the input port of the first second-order nonlinear element 804 to the first PPLN waveguide 822, and the second spatial optical system 823 includes the first PPLN waveguide. The light output from 822 is coupled to the output port of the first second-order nonlinear element 804 via the dichroic mirror 824. The third spatial optical system 825 couples light input from the input port of the second second-order nonlinear element 810 to the second PPLN waveguide 826, and the fourth spatial optical system 827 includes the second PPLN waveguide. The light output from 826 is coupled to the output port of the second second-order nonlinear element 810 via the dichroic mirror 828. The fifth spatial optical system 829 couples the light input from the input port of the third second-order nonlinear element 811 to the third PPLN waveguide 830 via the dichroic mirror 832, and the sixth spatial optical system 831 The light output from the third PPLN waveguide 830 is coupled to the output port of the third second-order nonlinear element 811 via the dichroic mirror 833. The seventh spatial optical system 834 couples the light input from the input port of the fourth second-order nonlinear element 814 to the fourth PPLN waveguide 835, and the eighth spatial optical system 836 includes the fourth PPLN waveguide. The light output from 835 is coupled to the output port of the fourth second-order nonlinear element 814 via the dichroic mirror 837. The ninth spatial optical system 838 couples the light input from the input port of the fifth second-order nonlinear element 815 to the fifth PPLN waveguide 839 via the dichroic mirror 841, and the tenth spatial optical system 840 The light output from the fifth PPLN waveguide 839 is coupled to the output port of the fifth second-order nonlinear element 815 via the dichroic mirror 842.

  In the example shown in FIG. 8, the signal light 852 incident on the optical amplifying apparatus 800, for example, QPSK modulated in the 1.55 μm band, is branched by the first optical multiplexer / demultiplexer 830-1, and one of the lights is subjected to delay control. After passing through the optical device 820, the light enters the third second-order nonlinear optical element 811 and the other light enters the third optical multiplexer / demultiplexer 803-3. The first fundamental wave light 801 generated by the first fundamental wave light source 802 is branched by the second optical multiplexer / demultiplexer 803-2, and one light is incident on the fourth optical multiplexer / demultiplexer 803-4, The other is incident on the third optical multiplexer / demultiplexer 803-3. A part of the signal light 852 incident on the third optical multiplexer / demultiplexer 803-3 and a part of the first fundamental wave light 851 are multiplexed by the third optical multiplexer / demultiplexer 803-3, Is incident on the first EDFA 801, amplified, and incident on the first second-order nonlinear optical element 804.

In the first second-order nonlinear optical element 804, a first PPLN waveguide 822 having three pseudo phase matching wavelengths with a frequency interval of 100 GHz in a 1.55 μm band is manufactured. A method for manufacturing the first PPLN waveguide 822 is illustrated below. First, a periodic electrode subjected to spatial phase modulation having a period of 17 μm and a phase modulation period of 16.5 mm was formed on LiNbO ₃ to which Zn was added. Next, a polarization inversion grating corresponding to the above electrode pattern was formed in Zn: LiNbO ₃ by an electric field application method. Next, the Zn: LiNbO ₃ substrate having this periodically poled structure was directly bonded onto LiTaO _{3 serving} as a cladding, and heat treatment was performed at 500 ° C. to firmly bond both substrates. Next, the core layer was thinned to about 5 μm by polishing, and a ridge type optical waveguide was formed using a dry etching process. The length of the waveguide was 49.5 mm, which is three times the periodic phase modulation period. This waveguide can be adjusted in temperature by a Peltier element, and is a module capable of inputting and outputting light with a polarization maintaining fiber in a 1.55 μm band.

  Here, the wavelength conversion process in the first PPLN waveguide 822 will be described. FIG. 9A shows the frequency arrangement and phase matching characteristics in the first PPLN waveguide 822. As shown in FIG. 9A, the peak on the lowest frequency side among the three phase matching peaks of the first PPLN waveguide 822 is matched with the frequency of the signal light 852. Then, the wavelength of the first fundamental wave light 851 is set so as to be detuned to the lower frequency side by 100 GHz than the signal light 852. Under the above conditions, in the first PPLN waveguide 822, the signal light 852 is first converted into SH light, and the first fundamental wave light 851 is generated by difference frequency generation (DFG) using the SH light as excitation light. The first idler light 853 is converted. Since the frequency of the converted first idler light 853 matches the second phase matching wavelength of the first PPLN waveguide 822, the first idler light 853 is converted into SH light, and the SH light is converted into excitation light. As a result, the signal light is converted into the second idler light 854. Since the frequency of the second idler light 854 matches the third phase matching wavelength of the first PPLN waveguide, the second idler light 854 is converted into SH light, and the first light is pumped as the first light. The idler light 853 is converted into the third idler light 855. At this time, the phases of the first to third idler lights are given by the following equations.

φ _i1 = 2φ _s −φ _LO1 (Formula 12)
φ _i2 = 2φ _i1 −φ _s = 3φ _s −2φ _LO1 (Formula 13)
φ _i3 = 2φ _i2 −φ _i1 = 4φ _s -3φ _LO1 (Formula 14)

Here, φ _s is the phase of the signal light, φ _LO1 is the phase of the first fundamental wave light, φ _i1 is the phase of the first idler light, φ _i2 is the phase of the second idler light, and φ _i3 is the third phase. It represents the phase of idler light.

  The output of the first second-order nonlinear optical element 804 having the first PPLN waveguide 822 is input to the duplexer 805. The demultiplexer 805 has a function of separating and outputting only the third idler light. In the demultiplexer 805, only the third idler light 855 is separated and output and enters the optical circulator 806. Although a duplexer is used here, only the third idler light may be cut out using a band-pass filter. The third idler light 855 is injection-synchronized with the laser output from the second fundamental wave light source (semiconductor laser) 807 in the optical circulator 806, and the remaining intensity modulation component is removed and output as the second fundamental wave light 856. The

  The first fundamental wave light 851 and the second fundamental wave light 856 are multiplexed by the fourth optical multiplexer / demultiplexer 803-4 and then branched into two. One of the lights branched by the fourth optical multiplexer / demultiplexer 803-4 is amplified by the second EDFA 808 and then passes through only the first fundamental wave light 851 and the second fundamental wave light 856. After excessive spontaneous emission light is removed by the filter 809, the light is incident on the second PPLN waveguide 826 of the second second-order nonlinear optical element 810.

The manufacturing method of the second PPLN waveguide 826 is substantially the same as the manufacturing method of the first PPLN waveguide 822, but the second PPLN waveguide 826 has no phase modulation applied to the domain-inverted structure with a period of 17 μm. Therefore, the phase matching wavelength when evaluated by the second harmonic generation (SHG) is the frequency arrangement in the second PPLN waveguide 826 of the second second-order nonlinear optical element 810 shown in FIG. 9B. Thus, it matches the wavelength of the first idler light 853. The first fundamental wave light 851 and the second fundamental wave light 856 are separated from each other in the opposite direction by 200 GHz from the phase matching wavelength. Therefore, the phase matching condition for the sum frequency generation of the first fundamental wave light 851 and the second fundamental wave light 856 is satisfied, and the first fundamental wave light 851 and the second fundamental wave light 856 are the second PPLN waveguide. Wavelength conversion is performed by the waveguide 826 to a wavelength corresponding to the SH light of the first idler light 853, and the light is output as the first sum frequency light 857. The phase φ _SF of the first sum frequency light 857 is given by the following (formula 15). Here, φ _LO2 indicates the phase of the second fundamental wave light.
φ _SF = φ _LO1 + φ _LO2 = 4φ _s -2φ _LO1 (Formula 15)

  The first sum frequency light 857 was separated into the first fundamental light 851 and the second fundamental light 856 by the first dichroic mirror 828 built in the second second-order nonlinear optical element 826. Later, the light enters the third PPLN waveguide 830 of the third second-order nonlinear optical element 811 through the polarization maintaining optical fiber in the 0.78 μm band.

  On the other hand, the signal light 852 branched by the first optical multiplexer / demultiplexer 803-1 also enters the third PPLN waveguide 830 via the third dichroic mirror 832. The third PPLN waveguide 830 has characteristics equivalent to those of the second PPLN waveguide 826, and therefore the phase matching wavelength thereof is substantially the same as that of the second PPLN waveguide 826. It is set to match by appropriate temperature adjustment.

FIG. 9C shows the frequency arrangement in the third PPLN waveguide. As shown in FIG. 9C, the frequency of the first sum frequency light 858 corresponds to the SH light having the phase matching wavelength of the third PPLN waveguide 830 in the 1.55 μm band. Therefore, the phase matching condition is satisfied between the first sum frequency light 858 and the signal light 852, and difference frequency (DFG) light is generated. The phase φ _i4 of the difference frequency light is expressed as shown in (Equation 16) below.
φ _i4 = 4φ _s -2φ _LO1 -φ ' _s (Formula 16)

Here, φ ′ _s is branched by the first optical multiplexer / demultiplexer 803-1, passes through the delay controller 820, and then enters the third PPLN waveguide 830 in the third second-order nonlinear optical element 811. The phase of the signal light 852 thus obtained. Usually, since the delay is different between the signal light 852 that is branched by the first multiplexer / demultiplexer 803-1 and passes through another path and the first sum frequency light 858, φ _s and φ _'s can not add subtraction on the formula. Therefore, by using the delay controller 820, the optical path length until one light branched from the first multiplexer / demultiplexer 803-1 enters the third PPLN waveguide 830 and the other branched light are used. The equal length was made so as to compensate for an effective difference in optical path length until the light enters the third PPLN waveguide 830. At this time, φ _s and φ ′ _s can be added and subtracted in a mathematical formula, and the phase φ _i4 of the difference frequency light is expressed by the following formula.
φ _i4 = 3φ _s -2φ _LO1 (Formula 17)

Comparing (Equation 13) with (Equation 17), it can be seen that this difference frequency light has a phase equivalent to the second idler light 854 generated by the first PPLN waveguide 822. That is, by using the third idler light 855 having a phase of 4φ _s , light having a phase equivalent to that of the second idler light 854 can be generated in a single difference frequency generation process. Here, this is referred to as fourth idler light 858. The signal light 852 and the fourth idler light 858 are emitted from the third second-order nonlinear optical element 811 via the fourth dichroic mirror 833.

  FIG. 10 shows the optical spectrum of the output of the third PPLN waveguide. As shown in FIG. 10, the fourth idler light 858 is generated as a result of the 10 Gbit / s QPSK signal light 852 and the sum frequency light 857 entering the third PPLN waveguide 830. Regarding the fourth idler light 858, it was confirmed that the intensity difference from the QPSK signal light 852 was generated within 3 dB.

On the other hand, after the other of the lights branched by the fourth optical multiplexer / demultiplexer 803-4 is amplified by the third EDFA 812, only the first fundamental wave light 851 and the second fundamental wave light 856 are used. After excessive spontaneous emission light is removed by the band-pass filter 813 that passes through, the light is incident on the fourth PPLN waveguide 835 of the fourth second-order nonlinear optical element 814. The fourth PPLN waveguide 835 has characteristics equivalent to those of the second PPLN waveguide 826, and therefore the phase matching wavelength thereof is almost the same as that of the second PPLN waveguide 826, and the phase matching wavelengths of both are It is set to match by appropriate temperature adjustment. The second sum frequency light 859 is generated in the same manner as the first sum frequency light 857 generated by the second PPLN waveguide 826. At this time, the phase φ _SF of the second sum frequency light 859 is similarly given by (Equation 15).

  The second sum frequency light 859 generated by the fourth PPLN waveguide 835 is emitted from the fourth second-order nonlinear optical element 814 via the fifth dichroic mirror 837, and the fifth second-order nonlinear optical element. 815 is incident. Further, the signal light 852 and the fourth idler light 858 emitted from the third second-order nonlinear optical element 811 are incident on the fifth second-order nonlinear optical element 815. The signal light 852, the fourth idler light 858, and the second sum frequency light 859 are incident on the fifth PPLN waveguide 839 via the sixth dichroic mirror 841. The fifth PPLN waveguide 839 has the same characteristics as the third PPLN waveguide 830, and therefore the phase matching wavelength thereof is almost the same as that of the third PPLN waveguide 830. It is set to match by appropriate temperature adjustment.

When the sum of the phase of the signal light 852 and the phase of the fourth idler light 858 is calculated, it is expressed as (Equation 18) below.
φ _s + φ _i4 = 4φ _s -2φ _LO1 (Formula 18)

In the phase sensitive amplification, when φ _pump −φ _signal −φ _idler becomes π / 2, maximum amplification is obtained when −p / 2, so that maximum attenuation is obtained. In the case of a pair, it can be seen that if the phase of the pumping light is taken as a reference, the signal phase is matched only when it is an integer multiple of π / 2, so that the QPSK signal can be phase-sensitively amplified.

  FIG. 11 shows optical spectra of the signal light 852 and the fourth idler light 858 amplified by the optical amplifying apparatus 800 according to the first embodiment. Unlike the conventional PSA using an optical fiber, there is no strong excitation light in the 1.55 μm band as shown in FIG. 11, and spontaneous emission light generated by the EDFA used for generating the excitation light is also mixed. Therefore, it can be confirmed that phase sensitive amplification can be performed while maintaining a high S / N ratio.

  In the actual amplification operation, in order to suppress the influence of the phase fluctuation due to the fluctuation of the optical path length due to the expansion and contraction of the optical fiber connecting each optical component, the first embodiment amplifies the output from the output of the fifth PPLN waveguide 839. Only the signal light that has been output is extracted using the bandpass filter 816, and a part of the signal light is branched using the fifth optical multiplexer / demultiplexer 803-5, detected by the photodetector 817, and then the PLL circuit 818 is used. Through this, feedback is provided to an optical fiber expander 819 comprising a phase modulator / PZT to realize stable operation. The position of the PZT is a position at which the phase between the signal light / idler light pair and the excitation light in the PPLN waveguide (the fifth PPLN waveguide 839 in the first embodiment) that performs optical parametric amplification can be adjusted. I just need it. In this embodiment, the phase fluctuation is suppressed by changing the phase of the fundamental wave light and adjusting the phase of the excitation light. However, for example, the position where the phase of the signal light and the idler light is changed, An arrangement that directly changes the phase of light may be used.

  In the present invention, unlike a phase sensitive amplifier using a conventional optical fiber, a mechanism such as phase modulation for avoiding stimulated Brillouin scattering is not required, and a nonlinear medium can be a waveguide of about 50 mm at most. As a whole, a compact phase sensitive amplifier can be constructed. FIG. 12 shows a constellation map of an input signal and an output signal when a QPSK signal intentionally added with phase noise is input to the phase sensitive optical amplifying apparatus 800 according to the first embodiment. As shown in FIG. 12, in the phase sensitive optical amplifying apparatus 800 according to the first embodiment, amplification is performed only when the phase of the signal is an integral multiple of π / 2. It can be seen that the phase reproduction type amplification is performed.

  In the first embodiment, the first fundamental wave light 851 is arranged on the lower frequency side than the signal light 852, but it goes without saying that the same effect can be obtained even if this arrangement is arranged in the reverse order. .

In the first embodiment, the first PPLN waveguide 822 has three phase matching wavelengths, but in order to obtain the same effect, the phase matching wavelength of the first PPLN waveguide 822 is changed from the low frequency side. Two may be used. In this case, the second idler light 854 does not work as excitation light, but when the first fundamental wave light 851 is converted by DFG using the SH light of the first idler light 853 as excitation light, the second idler light 854 in FIG. 3 idler light having the following phase φ _{i3 ′} is generated at the same frequency as the idler light 855 of 3.
φ _{i3 ′} = 2φ _i1 −φ _LO1 = 4φ _s −3φ _LO1 (Formula 19)

As is clear from the comparison of (Equation 15) and (Equation 19), it can be seen that the phases of both are the same, and the third phase matching peak is not necessarily essential. Therefore, in order to obtain the same effect as in the first embodiment, the quasi phase matching wavelength of the first PPLN waveguide 822 is set to the frequency ω _{LO1 of} the first fundamental wave light 851 and the frequency ω _s of the signal light 852. On the other hand, it is only necessary to match the frequency of at least ω _s , 2ω _s −ω _LO1 .

(Example 2)
Hereinafter, an optical amplification apparatus according to Embodiment 2 of the present invention will be described. In the first embodiment, the phase reproduction amplification of the QPSK signal is performed using the PPLN waveguide having two to three phase matching wavelengths. However, the second embodiment further supports the 8PSK signal which is a multilevel signal format. A PPLN waveguide was configured as described above. The optical amplifying device according to the second embodiment is characterized in that a carrier wave is extracted from a modulation signal having an eight-level phase. Further, in the first embodiment, the fourth idler light is generated from the signal light by the difference frequency generation on the main signal line, but in the second embodiment, the idler light that makes a pair with the signal light is separated from the main signal line. It is generated with. As a result, it is not necessary to use a non-linear process in the previous stage that is incident on the optical parametric amplifier, so that an optical amplifier having better noise characteristics can be configured.

  FIG. 13 shows a configuration of an optical amplifying device according to Embodiment 2 of the present invention. 13 includes first to third EDFAs 1301, 1308, and 1313, first and second fundamental wave light sources 1302 and 1307, first to sixth optical multiplexers 1303-1 to 1303-6, First to fifth second-order nonlinear optical elements 1304, 1310, 1311, 1314, 1315, a demultiplexer 1305, an optical circulator 1306, first to fourth bandpass filters 1309, 1313, 1316, 1343 , An optical amplifying apparatus 1300 including a photodetector 1317, a PLL circuit 1318, an optical fiber stretcher 1319 by PZT, first and second delay controllers 1320 and 1344, and a wavelength multiplexer 1345 is shown. Has been.

  The first second-order nonlinear optical element 1304 includes first and second spatial optical systems 1321 and 1323, a first PPLN waveguide 1322, and a first dichroic mirror 1324. The second second-order nonlinear optical element 1310 includes third and fourth spatial optical systems 1325 and 1327, a second PPLN waveguide 1326, and a second dichroic mirror 1328. The third second-order nonlinear optical element 1311 includes fifth and sixth spatial optical systems 1329 and 1331, a third PPLN waveguide 1330, and third and fourth dichroic mirrors 1323 and 1333. The fourth second-order nonlinear optical element 1314 includes seventh and eighth spatial optical systems 1334 and 1336, a fourth PPLN waveguide 1335, and a fifth dichroic mirror 1337. The fifth second-order nonlinear optical element 1315 includes ninth and tenth spatial optical systems 1338 and 1340, a fifth PPLN waveguide 1339, and sixth and seventh dichroic mirrors 1341 and 1342.

  As shown in FIG. 13, the overall configuration of the optical amplifying apparatus 1300 according to the second embodiment is the same as that of the optical amplifying apparatus 800 according to the first embodiment shown in FIG. −6, the second delay controller 1344, the fourth bandpass filter 1343, and the wavelength multiplexer 1345 are increased, but since they are almost the same, the details thereof are assumed to have the same functions unless otherwise noted. Such description is omitted.

  First, carrier wave extraction using the first PPLN waveguide 1322 according to the optical amplification device 1300 of the second embodiment will be described. As shown in FIG. 13, the signal light and the first fundamental wave light are amplified by the first EDFA 1301 and then enter the first PPLN waveguide 1322 having a plurality of phase matching wavelengths. The first fundamental wave light is set to be detuned to the lower frequency side by 100 GHz than the signal light. The first PPLN waveguide 1322 is designed so as to have a phase matching peak at three wavelengths: a signal light wavelength, a wavelength far from the 100 GHz high frequency side, and a wavelength far from the 300 GHz high frequency side. For example, the first PPLN waveguide 1322 is subjected to spatial phase modulation with a period of 16.92 μm and a phase modulation period of 16.5 mm.

A wavelength conversion process in the first PPLN waveguide 1322 will be described. In the first PPLN waveguide 1322, the signal light is first converted into SH light, and the first fundamental light is converted into first idler light by difference frequency generation (DFG) using the SH light as excitation light. . Since the frequency of the converted first idler light matches the second phase matching wavelength of the first PPLN waveguide 1322, the first idler light is converted into SH light, and the SH light is used as excitation light. The signal light is converted into second idler light. Furthermore, when the first fundamental wave light is converted by the DFG using the SH light of the first idler light as excitation light, third idler light is generated. Since the frequency of the converted first idler light matches the second phase matching wavelength of the first PPLN waveguide 1322, the third idler light is converted into SH light, and the SH light is used as excitation light. The second idler light to the fourth idler light, the first idler light to the fifth idler light, the signal light to the sixth idler light, and the first fundamental light to the seventh idler light, respectively. Converted. At this time, when attention is paid to the phases of the first idler light, the third idler light, the sixth idler light, and the seventh idler light, the phases of these idler lights are expressed by the following (Expression 20) to (Expression). 23).
φ _i1 = 2φ _s −φ _LO1 (Formula 20)
φ _i3 = 2φ _i2 −φ _LO1 = 4φ _s -3φ _LO1 (Formula 21)
φ _i6 = 2φ _i3 −φ _s = 7φ _s −6φ _LO1 (Formula 22)
φ _i7 = 2φ _i3 −φ _LO1 = 8φ _s −7φ _LO1 (Equation 23)

Here, φ _i6 indicates the phase of the sixth idler light, and φ _i7 indicates the phase of the seventh idler light. As described above, the first to seventh idler lights can be sequentially generated only by setting the three phase matching wavelengths. That is, the quasi phase matching wavelength of the first PPLN waveguide 1320 is at least ω _s , 2ω _s −ω _LO1 , 4ω _s with respect to the frequency ω _{LO1 of} the first fundamental wave light and the frequency ω _{s of the} signal light. It only needs to match the frequency of −3ω _LO1 .

  Paying attention to the phase of the seventh idler light expressed by (Equation 23), since the phase of the signal light is multiplied by 8, the signal having the phase that is an integer multiple of π / 4 of 8PSK is the first. It can be seen that the phase of the idler light 7 becomes constant and the carrier wave can be extracted.

Refer to FIG. 13 again. As in the first embodiment, since the actual phase modulation signal is accompanied by a change in intensity, only the seventh idler light is extracted from the output of the first second-order nonlinear optical element 1304, and then light is injected by the optical circulator 1306. By performing the synchronization, the second fundamental wave light is obtained. Thereafter, a first sum frequency light between the first fundamental wave light and the second fundamental wave light is generated by the second second-order nonlinear optical element 1310. A third second-order nonlinear optical element 1311 uses the generated first sum frequency light and part of the signal light branched by the first and sixth multiplexers / demultiplexers 1303-1 and 1303-6. The eighth idler light is generated by generating the difference frequency. At this time, the phase of the eighth idler light is expressed by the following equation.
φ _i8 = 7φ _s -6φ _LO1 (Formula 24)

  As shown in (Equation 23), the phase of the eighth idler light is equivalent to the phase of the sixth idler light. The effective optical path length of the two paths branched by the first multiplexer / demultiplexer 1303-1 is adjusted by the first delay controller 1320.

  The difference from the configuration of the first embodiment is that idler light (fourth idler light in the first embodiment and eighth idler light in the second embodiment) that interacts with the signal in a pair in the optical parametric amplification process on the main signal line. There is that it is not arranged. In the first embodiment, the fourth idler light is generated from the main signal, and the pair of the signal light and the idler light is incident on the fifth second-order nonlinear optical element. However, the light is generated by the noise generated in the difference frequency generation process. Degrading the characteristics of the amplifier. For this reason, in Example 2, in order to further improve the noise characteristics, the generation process of the eighth idler light is performed through a path different from the main signal line.

  Of the output of the third second-order nonlinear optical element 1311, only the eighth idler light is extracted using the fourth bandpass filter 1343, and the signal light that has passed through the main signal line by the wavelength multiplexer / demultiplexer 1345. And combined. Here, of the two lights branched by the sixth multiplexer / demultiplexer 1303-6, the optical path length until the light enters the wavelength multiplexer / demultiplexer 1345 after passing through the second delay controller 1344, After passing through the first multiplexer / demultiplexer 1303-1, the first delay controller 1320, the third second-order nonlinear optical element 1311, and the fourth bandpass filter 1343, the light enters the wavelength multiplexer / demultiplexer 1345. The optical path length up to is completely equalized by the second delay controller 1344. The signal light combined with the wavelength multiplexer / demultiplexer 1345 and the eighth idler light are incident on the fifth second-order nonlinear optical element 1315.

  On the other hand, the other first fundamental wave light and second fundamental wave light branched by the fourth multiplexer / demultiplexer 1303-4 are amplified by the EDFA 1312 and then the second second fundamental wave light, as in the first embodiment. The light enters the next nonlinear optical element 1314 to generate second sum frequency light. This second sum frequency light is also incident on the fifth second-order nonlinear optical element 1315.

  In the optical parametric amplification process using the second sum frequency light, the signal light, and the eighth idler light, the signal light is amplified only when the phase of the signal light is an integral multiple of π / 4. Therefore, the 8PSK signal is phase-sensitive amplified. be able to. FIG. 14 shows a constellation map of an input signal and an output signal when a QPSK signal intentionally added with phase noise is input to the phase sensitive amplifier according to the second embodiment. As shown in FIG. 14, in the second embodiment, amplification is performed only when the phase of the signal is an integer multiple of π / 4, so that the phase noise observed in the input signal can be reduced. It can be seen that phase reproduction type amplification is performed.

Here, when the configuration of the present invention is applied to an optical amplifier that amplifies multilevel phase-modulated signal light that takes M (M is an integer of 4 or more) kinds of values, the first second-order nonlinear optics is used. Only the idler light having the frequency of Mω _s − (M−1) ω _LO1 is extracted from the output of the element by the duplexer, and newly (M−1) ω _s − ( M-2) It is only necessary to generate idler light having a frequency of ω _LO1 and perform optical parametric amplification with the signal light.

PSA 100, 400
Phase sensitive optical amplifier 101
Excitation light source 102
Excitation light phase controller 103
1st optical branching part 104-1, 203-1
Second optical branching sections 104-2 and 203-2
EDFA 201
First second-order nonlinear optical element 202, 604, 804, 1304
Second second-order nonlinear optical element 204, 610, 810, 1310
Phase modulator 205
Optical fiber stretcher 206, 413, 615, 819, 1319 by PZT
Polarization maintaining fiber 207, 616
Photodetector 208, 411, 613, 817, 1317
Phase-locked loop circuit (PLL) 209, 412, 614, 818, 1318
First spatial optical system 211, 621, 821, 1321
First PPLN waveguide 212, 622, 822, 1322
Second spatial optical system 213, 623, 823, 1323
First dichroic mirror 214, 624, 824, 1324
Third spatial optical system 215, 625, 825, 1325
Second PPLN waveguides 216, 626, 826, 1326
Fourth spatial optical system 217, 627, 827, 1327
Second dichroic mirror 218, 628, 828, 1328
Third dichroic mirror 219, 632, 832, 1332
First EDFA 401, 601, 801, 1301
First excitation light source 402, 602
First optical multiplexer / demultiplexer 403-1, 603-1, 803-1, 1303-1
Second optical multiplexer / demultiplexer 403-2, 603-2, 803-2, 1303-2
Third optical multiplexer / demultiplexer 403-3, 603-3, 803-3, 1303-3
Fourth optical multiplexer / demultiplexer 403-4, 603-4, 803-4, 1303-4
Fifth optical multiplexer / demultiplexer 403-5, 803-5, 1303-5
First optical fiber 404
Demultiplexer 405, 605, 805, 1305
Optical circulator 406, 606, 806, 1306
Semiconductor lasers 407, 607, 807, 1307
Second EDFA 408, 608, 808, 1308
Second optical fiber 409
Band pass filter 410
Optical amplification device 600, 800, 1300
First band pass filter 609, 809, 1309
Third second-order nonlinear optical elements 611, 811 and 1311
Second band pass filter 612, 813, 1313
Fifth spatial optical system 629, 829, 1329
Third PPLN waveguide 630, 830, 1330
Sixth spatial optical system 631, 831, 1331
Fourth dichroic mirror 633, 833, 1333
First fundamental wave light source 802
Third EDFA 812, 1312
Fourth second-order nonlinear optical elements 814 and 1314
Fifth second-order nonlinear optical elements 815 and 1315
Third band pass filters 816, 1316
Delay controller 820, 1320, 1344
Seventh spatial optical system 834, 1334
Fourth PPLN waveguides 835, 1335
Eighth spatial optical system 836, 1336
Fifth dichroic mirror 837, 1337
Ninth spatial optical system 838, 1338
Fifth PPLN waveguide 839, 1339
Tenth spatial optical system 840, 1340
Sixth dichroic mirror 841, 1341
Seventh dichroic mirror 842, 1342
Sixth optical multiplexer / demultiplexer 1303-6
Fourth band pass filter 1343
Wavelength multiplexer 1345

Claims (7)

A phase-sensitive optical amplifying device that amplifies multilevel phase-modulated signal light that takes M (M is an integer of 4 or more) kinds of light mixing using a second-order nonlinear optical effect,
A first optical multiplexer / demultiplexer for branching the signal light having a frequency ω _s ;
A first light source for generating a first fundamental light having a frequency ω _LO1 ;
A second optical multiplexer / demultiplexer for branching the first fundamental wave light;
The signal light output from one output terminal of the first optical multiplexer / demultiplexer and the first fundamental wave light output from one of the second optical multiplexer / demultiplexers are input and amplified. A first optical fiber laser amplifier for outputting;
A first second-order nonlinear optical element that satisfies quasi-phase matching at a plurality of wavelengths, inputs the amplified signal light and the first fundamental light, generates a second harmonic, and outputs the second harmonic;
A multiplexer / demultiplexer for separating and outputting only idler light having a frequency of Mω _s − (M−1) ω _LO1 from the output of the first second-order nonlinear optical element;
A second light source that outputs a laser capable of injection locking to the idler light output from the duplexer;
An optical circulator that outputs the second fundamental wave light in synchronization with injection of the idler light output from the duplexer into the laser output from the second excitation light source;
The first fundamental wave light output from the other output terminal of the second optical multiplexer / demultiplexer and the second fundamental wave light output from the optical circulator are input and combined, and the combined signals are combined. A third optical multiplexer / demultiplexer for branching and outputting the first fundamental wave light and the second fundamental wave light;
A second optical fiber laser amplifier that amplifies and outputs the first fundamental wave light and the second fundamental wave light output from one output terminal of the third optical multiplexer / demultiplexer;
A second second-order nonlinear optical element that generates and outputs a first sum frequency light from the first fundamental wave light and the second fundamental wave light output from the second optical fiber laser amplifier;
The signal light output from the other output terminal of the first optical multiplexer / demultiplexer and the first sum frequency light are input to generate a difference frequency light, and the signal light and the difference frequency light are output. A third second-order nonlinear optical element,
A third optical fiber laser amplifier that amplifies and outputs the first fundamental wave light and the second fundamental wave light output from the other output terminal of the third optical multiplexer / demultiplexer;
A fourth second-order nonlinear optical element that generates and outputs a second sum frequency light from the first fundamental wave light and the second fundamental wave light output from the third optical fiber laser amplifier;
An optical amplification apparatus comprising: a fifth second-order nonlinear optical element that inputs the signal light, the difference frequency light, and the second sum frequency light and performs non-degenerate parametric amplification.   A first bandpass filter that transmits only the first fundamental wave light and the second fundamental wave light is provided between the second optical fiber laser amplifier and the second second-order nonlinear optical element. The optical amplifying device according to claim 1, wherein:   An optical path from the one output terminal of the first optical multiplexer / demultiplexer to the third secondary nonlinear optical element via the second secondary nonlinear optical element, and the first optical multiplexer / demultiplexer 3. The optical amplification according to claim 1, further comprising a first delay controller that equalizes an optical path from the other output terminal to the third second-order nonlinear optical element. apparatus. The first second-order nonlinear optical element is an optical waveguide made of LiNbO ₃ having a periodically poled structure,
The periodic polarization reversal structure is subjected to spatial phase modulation or periodic modulation with a period longer than the polarization reversal period,
The first second-order nonlinear optical element has a quasi-phase matching wavelength that matches at least the frequency of ω _s , 2ω _s -ω _LO1 , and amplifies a QPSK signal. The optical amplification device according to any one of the above. A fourth optical multiplexer / demultiplexer for branching the signal light and outputting the signal light from one output terminal to the first optical multiplexer / demultiplexer;
A second delay controller that inputs the signal light output from the other output terminal of the fourth optical multiplexer / demultiplexer and delays the signal light;
A second bandpass filter that transmits only the difference frequency light out of the signal light and the difference frequency light output from the third second-order nonlinear optical element;
The signal light output from the second delay control circuit and the difference frequency light output from the second bandpass filter are combined and output to the fifth second-order nonlinear optical element. The optical amplifier according to claim 1, further comprising: a wavelength multiplexer.   The second delay controller transmits the wavelength combining from the one output terminal of the fourth optical multiplexer / demultiplexer via the second second-order nonlinear optical element and the third second-order nonlinear optical element. The optical path to the optical multiplexer and the optical path from the other output terminal of the fourth optical multiplexer / demultiplexer to the wavelength multiplexer are configured to be equal in length. The optical amplifying device described. The first second-order nonlinear optical element is an optical waveguide made of LiNbO ₃ having a periodically poled structure,
The periodic polarization reversal structure is subjected to spatial phase modulation or periodic modulation with a period longer than the polarization reversal period,
The first second-order nonlinear optical element has a quasi phase matching wavelength that matches at least the frequencies of ω _s , 2ω _s −ω _LO1 , 4ω _s −3ω _LO1 , and amplifies an 8PSK signal. The optical amplification device according to claim 5 or 6.