Control Circuit and Optical Circuit Control Method

Abstract

This control circuit controls at least one optical path difference among optical path differences each between two optical paths included in one or more MZIs included in an optical circuit. This control circuit has: reference light beam generation units (35), a reference light beam input unit (8); optical power monitoring units (52, 53); and a controller (41). The reference light beam generation units generate, in parallel or in a time-division manner, reference light beams in N (N is a natural number) types of wavelengths different from the wavelength λ of an input light beam inputted to the optical circuit. The reference light beam input unit inputs the reference light beams to the optical circuit. The optical power monitoring units monitor the optical intensity of post-propagation input light beam that is the input light beam after having propagated through the optical circuit, the light intensities of post-propagation reference light beams that are the reference light beams after having propagated through the optical circuit, or the optical intensity of a light beam in which the post-propagation input light beam and the post-propagation reference light beams are multiplexed. The controller performs control such that then optical path difference is corrected on the basis of the optical intensity of the post-propagation input light beam and the light intensities of the N types of post-propagation reference light beams, obtained using the monitoring result.

Description

Technical Field

[0001] The present invention relates to a control circuit and an optical circuit control method. This application claims priority to PCT / JP2021 / 044767 filed on December 6, 2021, the content of which is incorporated herein by reference.

Background Art

[0002] An optical circuit composed of a Mach-Zehnder interferometer (MZI) or a combination of MZIs is widely used as an optical filter or an optical modulator having periodicity. In particular, for an IQ optical modulator for generating an optical QAM (Quadrature Amplitude Modulation) signal, an MZ configured in a nested state I is widely used.

[0003] FIG. 10 is a diagram showing the configuration of the MZI91. The MZI91 is the most basic configuration of the MZI. The optical input port 1 inputs input light. The optical branching circuit 2 branches the input light input to the optical input port 1 into two. In many cases, this branching ratio is 1:1. However, it is not necessarily limited to this ratio. One of the two lights branched by the optical branching circuit 2 propagates through the first optical waveguide 3, and the other light propagates through the second optical waveguide 4. The optical multiplexing circuit 6 multiplexes the light that has propagated through the first optical waveguide 3 and the light that has propagated through the second optical waveguide 4. The optical multiplexing circuit 6 outputs the multiplexed light to the output port 7. The optical path length of the second optical waveguide 4 is finely adjusted by the bias voltage Vb applied by the bias application electrode 5. For this fine adjustment, the Pockels effect or the thermal expansion of the optical waveguide by a heater can be used.

[0004] Let the optical path difference between the optical path length of the first optical waveguide 3 and the optical path length of the second optical waveguide 4 be the optical path difference ΔL. The magnitude of the optical intensity of the output light output from the output port 7 changes depending on the wavelength λ of the input light and the optical path difference ΔL. However, the wavelength λ is affected by the refractive index of the optical waveguide. Also, the optical path length is affected by the circuit configuration of the optical branching circuit 2 and the optical multiplexing circuit 6.

[0005] In the present application, for simplicity, the wavelength λ is the wavelength when the light is propagating inside the optical waveguide, rather than the optical wavelength in vacuum. Also, in the present application, the optical path length is the value obtained by adding the optical phase fluctuations inside the optical branching circuit 2 and inside the optical multiplexing circuit 6. As an example, in the process of light propagating from the second optical waveguide 4 to the output port 7, when the optical phase is delayed by π / 2 (the unit is radian, the same hereinafter) inside the optical multiplexing circuit 6, this phase delay is regarded as an increase in the optical path length equivalent to λ / 4. This is because the wavelength λ corresponds to the optical phase 2π.

[0006] FIG. 11 is a diagram showing the relationship between the optical output intensity P(λ), which is the intensity of the output light of the MZI91 shown in FIG. 10, and the optical path difference ΔL. FIG. 11(a) shows the relationship before the bias drift occurs, and FIG. 11(b) shows the relationship after the bias drift occurs. The bias drift will be described later.

[0007] As is well known, the optical output intensity P(λ) from the MZI shows a sine wave response with respect to the optical path difference ΔL. The state of maximum output, the state of minimum output, and the intermediate states between them are often called the Peak point, Null point, and Quad point, respectively. Therefore, this notation is also used in the present application. The Peak point is obtained when the optical path difference ΔL is an even multiple (including 0 times) of λ / 2, and the Null point is obtained when the optical path difference ΔL is an odd multiple of λ / 2. It should be noted here that generally, the Null point, Quad point, and Peak point depend on the wavelength, but only the Peak point at ΔL = 0 is exceptionally independent of the wavelength.

[0008] Consider using the MZI91 as an optical filter to suppress the light of wavelength λ. In this case, any one of the Null points for the wavelength λ can be selected. For example, by adjusting the bias voltage Vb applied by the bias application electrode 5, the optical path difference ΔL can be set to 0.5λ. In FIG. 11(a), this state is indicated by the arrow A9.

[0009] However, even if the bias voltage Vb is constantly maintained, the optical output intensity may change over time due to temperature changes outside the MZI91 or applied stress. For example, as shown in Fig. 11(a), even though the bias is set at the Null point and the optical path difference ΔL is 0.5λ, over time, as shown in Fig. 11(b), the optical path difference ΔL may become a value different from 0.5λ. This is called bias drift in the MZI.

[0010] In the case of a single MZI, the presence or absence of bias drift can be monitored relatively easily. If it is a drift from the Null point, the optical output intensity increases, and if it is a drift from the Peak point, the optical output intensity decreases. Therefore, bias drift can be detected by monitoring the optical output intensity.

[0011] However, even if it is confirmed that bias drift has occurred, it is not obvious whether the bias voltage Vb should be increased or decreased to compensate for this drift. Also, the optical output intensity of the light source is not completely constant and has a slight fluctuation. Therefore, there is also a problem that it is necessary to distinguish whether the fluctuation of the optical output intensity is due to bias drift or due to the fluctuation of the light source.

[0012] To solve these problems, a technique of slightly changing the bias voltage Vb and performing synchronous detection has already been proposed. In this technique, dithering is performed by superimposing a pilot tone with a small amplitude of frequency f on the bias voltage Vb, and the minute change in the optical output intensity P(λ) is synchronously detected at frequency f. Based on the positive or negative of the synchronous detection result, it becomes possible to determine whether the bias voltage Vb should be increased or decreased. In addition, circuit noise other than the frequency f is suppressed. Therefore, there is an advantage that monitoring of the bias condition can be achieved with high accuracy.

[0013] The above-mentioned simple method was related to an optical filter composed of a single MZI. On the other hand, by applying an external modulation signal to a single MZI to modulate the optical path difference ΔL and modulating the output light, it is also widely practiced to utilize a single MZI as an optical modulator. At this time, if it is biased so that the output light becomes the Quad point at the moment when the modulation signal is zero, an NRZ (Non Return to Zero) signal can be obtained. Alternatively, if it is biased so that the output light becomes the Null point at the moment when the modulation signal is zero, a CS-RZ (Carrier-Suppressed Return-to-Zero) signal can be obtained. Utilizing a single MZI as an optical modulator Even in this case, techniques for detecting and compensating for bias drift using synchronous detection have already been proposed.

[0014] Next, the case of generating a QAM signal using an IQ optical modulator composed of nested MZIs will be described. FIG. 12 is a diagram showing the configuration of the IQ optical modulator 92. The IQ optical modulator 92 is an IQ optical modulator with a typical configuration. The IQ optical modulator 92 has a configuration in which an In-Phase (in-phase) MZI 50 is installed in the first optical waveguide 3 of the MZI 91 shown in FIG. 10, and a Quadrature (quadrature phase) MZI 51 is installed in the second optical waveguide 4.

[0015] Inside the In-Phase MZI 50, the I-side optical branching circuit 20 branches the light transmitted through the first optical waveguide 3 into two, and outputs the branched light one by one to each of the two optical paths. The I-side optical multiplexing circuit 21 multiplexes these two branched lights. The optical path difference between the two optical paths inside the In-Phase MZI 50 is push-pull modulated by the modulation signal I applied via the I-side modulation electrode 22. Here, the In-Phase MZI 50 is biased by the bias voltage Vb_I applied via the I-side bias application electrode 23 so that it becomes the Null point at the moment when the modulation signal I is zero. As a result, the optical electric field E_I of the modulated light output from the In-Phase MZI 50 has two types of optical phases that differ by π randomly.

[0016] Inside the MZI51 for Quadrature, the Q-side optical branching circuit 24 branches the light transmitted through the second optical waveguide 4 into two, and outputs the branched light one by one to each of the two optical paths. The Q-side optical multiplexing circuit 25 multiplexes these two branched lights. The optical path difference between the two optical paths inside the MZI51 for Quadrature is push-pull modulated by the modulation signal Q applied via the Q-side modulation electrode 26. Here, the MZI51 for Quadrature is biased by the bias voltage Vb_Q applied via the Q-side bias application electrode 27 so as to become a null point at the moment when the modulation signal Q is zero. As a result, the optical electric field E_Q of the modulated light output from the MZI51 for Quadrature also randomly has two types of optical phases different by π.

[0017] The optical multiplexing circuit 6 multiplexes the modulated light output from the MZI50 for In-Phase and the modulated light output from the MZI51 for Quadrature. The optical path difference between the two is adjusted by the bias voltage Vb applied via the bias application electrode 5. The bias voltage is set so that the optical phases of the optical electric field E_I and the optical electric field E_Q have a difference of ±π / 4 + π×m ph where m ph is an arbitrary integer. In other words, the largest-scale MZI composed of the first optical waveguide 3 and the second optical waveguide 4, which corresponds to the parent MZI in the nested MZI structure, is biased to the Quad point by the bias voltage Vb. As a result, the modulated light output from the optical multiplexing circuit 6 becomes a QAM signal. The output port 7 outputs the QAM signal.

[0018] In many IQ optical modulators, the power of the modulated light is monitored by an optical power monitor in order to monitor the conditions of each bias voltage. In FIG. 12, the optical tap circuit 10 taps the light output to the output port 7 and inputs the tapped light to the optical power monitor 11. Also, the optical multiplexing circuit 6 may be replaced with a 2-input 2-output coupler, and one of the two outputs from this coupler may be connected to the optical power monitor 11.

[0019] FIG. 13 is a diagram showing the optical electric field of the optical QAM signal output from the optical multiplexing circuit 6 of the IQ optical modulator 92 shown in FIG. 12 as a constellation. FIG. 13(a) shows the optical electric field before the bias drift occurs, and FIGS. 13(b) and 13(c) show the optical electric field after the bias drift occurs.

[0020] If all of the bias voltages Vb_I, Vb_Q, and Vb of the IQ optical modulator 92 are optimally set and the modulation signals I and Q of the drive signal have a constant amplitude, the generated modulation signal is a 4-value QAM, that is, a QPSK (Quadrature Phase Shift Keying) signal. FIG. 13(a) shows the optical electric field of the QAM signal in this case.

[0021] FIGS. 13(b) and 13(c) show the constellations when the bias voltage Vb_I and the bias voltage Vb_Q are appropriate but the bias voltage Vb deviates from the optimum value. Since the constellation is distorted, the signal quality is significantly deteriorated. However, each symbol of the constellation maintains a position symmetric with respect to the origin. Therefore, the optical intensity of the modulated light shown in each of FIGS. 13(b) and 13(c) is the same as the optical intensity of the modulated light shown in FIG. 13(a). This property does not change even for a constellation larger than 4 values, for example, 16-QAM. Therefore, when generating a QAM signal by an IQ optical modulator, determining whether the bias voltage Vb is optimal is much more difficult than determining whether the bias voltages Vb_I and Vb_Q are optimal.

[0022] However, a technical solution has already been proposed for this problem. One of them is a method called asymmetric bias dithering (see, for example, Non-Patent Document 1). Asymmetric bias dithering adds a dithering with a small amplitude whose phases are orthogonal to each of the bias voltages Vb_I and Vb_Q. This enables monitoring of all bias conditions of the bias voltage Vb_I, the bias voltage Vb_Q, and the bias voltage Vb.

Prior Art Documents

Non-Patent Literature

[0023]

Non-Patent Literature 1

Summary of the Invention

Problems to be Solved by the Invention

[0024] In the MZI control technology described so far, there has been a problem that a single or a plurality of bias voltages have to be intentionally varied slightly from the optimum point. In particular, when a pilot tone of frequency f is added to the bias voltage for dithering, noise of frequency f or an integer multiple thereof, although with a small amplitude, is superimposed on the output light of the MZI.

[0025] When the MZI is used as an optical filter, this noise has an adverse effect as a slight variation in the intensity of the output light or a slight variation in the wavelength transmission characteristics. Also, when the MZI is utilized as a modulator, this noise becomes a factor of slight signal quality degradation. In a high-speed large-capacity transmission system, even a slight signal quality degradation is difficult to tolerate, so the amplitude of dithering is strongly restricted. However, restricting the amplitude of dithering deteriorates the sensitivity of monitoring the bias condition. In some cases, there has been a problem that it causes constellation distortion and extremely serious degradation of signal quality.

[0026] In view of the above circumstances, an object of the present invention is to provide a control circuit and an optical circuit control method capable of adjusting the bias of an optical circuit composed of a single or a plurality of MZIs without adding a pilot tone to the bias voltage so as to improve the quality of the output from the optical circuit.

Means for Solving the Problems

[0027] One aspect of the present invention is a control circuit for controlling at least one of the optical path differences between two optical paths of each of a single or a plurality of Mach-Zehnder interferometers included in an optical circuit. When N is a natural number, N types of wavelengths λ1 to λ different from the wavelength λ of the input light input to the optical circuit N A reference light generation unit that generates each of the reference lights in parallel or in a time-division manner, a reference light input unit that inputs the reference light to the optical circuit, the optical intensity of the propagated input light that is the input light after propagating through the single or plurality of Mach-Zehnder interferometers included in the optical circuit, and the propagated input light A light power monitor unit that monitors the optical intensity of the propagated reference light that is the reference light after propagating through the single or plurality of Mach-Zehnder interferometers, or the optical intensity of the light obtained by multiplexing the propagated input light and the propagated reference light, and the optical intensity P(λ) of the propagated input light of wavelength λ obtained using the monitoring result by the light power monitor unit and N types of wavelengths λ1 to λ N The optical intensity P(λ1) to P(λ N ) of each of the propagated reference lights, and a controller that controls to correct the optical path difference based on the above.

[0028] One aspect of the present invention is a control circuit for controlling at least one of the optical path differences between two optical paths of each of a single or a plurality of Mach-Zehnder interferometers included in an optical circuit. When N is a natural number, N types of wavelengths λ1 to λ different from the wavelength λ of the input light input to the optical circuit NA reference light generation unit that generates each reference light in parallel or in a time-division manner, a reference light input unit that inputs the reference light into the optical circuit, the light intensity of the propagated input light that is the input light after propagating through the single or multiple Mach-Zehnder interferometers included in the optical circuit, and the light intensity of the propagated reference light that is the reference light after propagating through the single or multiple Mach-Zehnder interferometers through which the propagated input light has propagated, or a light power monitor unit that monitors the light intensity of the light obtained by multiplexing the propagated input light and the propagated reference light, and N types of wavelengths λ1 to λ obtained using the monitoring result by the light power monitor unit N Based on the light intensities P(λ1) to P(λ N ) of the respective propagated reference lights, a controller that controls to correct the optical path difference, is provided.

[0029] One aspect of the present invention is an optical circuit control method for controlling at least one of the optical path differences of two optical paths each included in a single or multiple Mach-Zehnder interferometers included in an optical circuit. When N is a natural number, N types of wavelengths λ1 to λ different from the wavelength λ of the input light input to the optical circuit N A reference light generation step of generating each reference light in parallel or in a time-division manner, a reference light input step of inputting the reference light into the optical circuit, the light intensity of the propagated input light that is the input light after propagating through the single or multiple Mach-Zehnder interferometers included in the optical circuit, and the light intensity of the propagated reference light that is the reference light after propagating through the single or multiple Mach-Zehnder interferometers through which the propagated input light has propagated, or the light intensity of the light obtained by multiplexing the propagated input light and the propagated reference light, a monitoring step of monitoring, and the light intensity P(λ) of the propagated input light of wavelength λ and N types of wavelengths λ1 to λ obtained using the monitoring result in the monitoring step N Based on the light intensities P(λ1) to P(λ N ) of the respective propagated reference lights, or a control step of controlling to correct the optical path difference based on the light intensities P(λ1) to P(λ N ) without using the light intensity P(λ), is included.

Advantages of the Invention

[0030] According to the present invention, it becomes possible to adjust the bias of an optical circuit composed of one or more MZIs so that the quality of the output from the optical circuit is improved without adding a pilot tone to the bias voltage.

Brief Description of the Drawings

[0031]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Modes for Carrying Out the Invention

[0032] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. This embodiment relates to a control circuit for precisely controlling the optical path length of a Mach-Zehnder interferometer (MZI).

[0033] <First Embodiment> FIG. 1 is a diagram showing the configuration of an optical output circuit 101 according to the first embodiment. Using FIG. 1, the MZI control and its operation according to the first embodiment will be described. In the optical output circuit 101 shown in FIG. 1, the same parts as those of the MZI 91 shown in FIG. 10 are denoted by the same reference numerals. In the optical output circuit 101, the optical circuit to be controlled by the MZI is an optical filter composed of a single MZI. This optical filter has many parts in common with the MZI 91 shown in FIG. 10.

[0034] One difference between the configuration of the optical filter included in the optical output circuit 101 shown in FIG. 1 and the configuration of the MZI 91 shown in FIG. 10 is that a wavelength division multiplexing coupler 8 is provided at the optical input port 1. The wavelength division multiplexing coupler 8 multiplexes the input light of wavelength λ input to the optical input port 1 and the reference lights output from each of the N (N is a natural number) reference light sources 35 and inputs the multiplexed light to the MZI. The nth (n is an integer from 1 to N) reference light source 35 is described as the nth reference light source 35-n, and the reference light of wavelength λ n output from the nth reference light source 35-n is described as the nth reference light. The wavelengths λ1 to λ N are different from each other. Hereinafter, N = 2 will be taken as an example for description. That is, the wavelength division multiplexing coupler 8 multiplexes the input light of wavelength λ input to the optical input port 1, the first reference light of wavelength λ1 output from the first reference light source 35-1, and the second reference light of wavelength λ2 output from the second reference light source 35-2 and inputs the multiplexed light to the MZI.

[0035] Here, it is also possible to use one type of reference light. However, in order to improve the control accuracy, it is desirable to have two or more types. In addition, the wavelength λ1 or wavelength λ2 of the reference light is preferably significantly different from the wavelength λ. Furthermore, it is more desirable that at least one of the plurality of reference lights has a longer wavelength than λ, and the other reference lights have a shorter wavelength than λ. However, the difference between the wavelength λ1 and the wavelength λ and the difference between the wavelength λ2 and the wavelength λ are preferably values different from the free spectral range. That is, it is desirable to avoid wavelength selection such that the wavelength λ and the wavelength λ1 become parallel Null (or parallel Peak), or the wavelength λ and the wavelength λ2 become parallel Null (or parallel Peak). In this embodiment, the description will be consistently made assuming λ2 > λ > λ1.

[0036] The optical branching circuit 2 branches the light multiplexed by the wavelength multiplexer 8 into two. One of the branched lights propagates through the first optical waveguide 3, and the other light propagates through the second optical waveguide 4. The optical path length of the second optical waveguide 4 is finely adjusted by the bias voltage Vb applied by the bias application electrode 5. The optical multiplexing circuit 6 multiplexes the light that has propagated through the first optical waveguide 3 and the light that has propagated through the second optical waveguide 4, and outputs the multiplexed light to the output port 7.

[0037] Another difference between the configuration of the optical filter included in the optical output circuit 101 shown in FIG. 1 and the configuration of the conventional MZI91 shown in FIG. 10 is that an optical bandpass filter 12 is provided at the output port 7. The optical bandpass filter 12 receives the light multiplexed by the optical multiplexing circuit 6, and blocks the first reference light of wavelength λ1 to the Nth reference light of wavelength λ N from the input light, and outputs only the output light of wavelength λ. When N = 2, the optical bandpass filter 12 blocks the first reference light of wavelength λ1 and the second reference light of wavelength λ2 from the input light, and outputs only the output light of wavelength λ. If the wavelength λ1 of the first reference light and the wavelength λ2 of the second reference light are significantly different from the wavelength λ, the transmission band of the optical bandpass filter 12 can be made sufficiently wide. That is, the accuracy required for setting the center wavelength of the transmission band is relaxed.

[0038] Another difference between the configuration of the optical filter included in the optical output circuit 101 shown in FIG. 1 and the configuration of the conventional MZI 91 shown in FIG. 10 is that it includes an optical tap circuit 10, a wavelength separation coupler 13, an optical power monitor 52, N reference optical power monitors 53, and a controller 41. The n-th reference optical power monitor 53 is denoted as the n-th reference optical power monitor 53-n.

[0039] The optical tap circuit 10 is provided between the optical multiplexing circuit 6 and the optical bandpass filter 12. The optical tap circuit 10 taps the output light of the optical multiplexing circuit 6 before the optical bandpass filter 12 blocks the reference light.

[0040] The wavelength separation coupler 13 separates the light tapped by the optical tap circuit 10 into the input light with wavelength λ, the first reference light with wavelength λ1 to the N-th reference light with wavelength λ N The wavelength separation coupler 13 is connected to the optical power monitor 52 and the first reference optical power monitor 53-1 to the N-th reference optical power monitor 53-N. The optical power monitor 52 monitors the input light with wavelength λ separated by the wavelength separation coupler 13. The n-th reference optical power monitor 53-n monitors the n-th reference light with wavelength λ n separated by the wavelength separation coupler 13.

[0041] The controller 41 determines whether the optical path difference between the two optical paths of the MZI included in the optical output circuit 101 is at an optimal value by comparing the light intensity of the light with wavelength λ, the light intensity of the first reference light with wavelength λ1,..., the light intensity of the N-th reference light with wavelength λ N If the controller 40 determines that it is not at the optimal value, it corrects the optical path difference by controlling the bias voltage Vb.

[0042] When N = 2, the wavelength separation coupler 13 separates the light tapped by the optical tap circuit 10 into input light of wavelength λ, first reference light of wavelength λ1, and second reference light of wavelength λ2. The optical power monitor 52 monitors the optical intensity of the input light of wavelength λ, the first reference light power monitor 53-1 monitors the optical intensity of the first reference light of wavelength λ1, and the second reference light power monitor 53-2 monitors the optical intensity of the second reference light of wavelength λ2. The controller 41 controls the bias voltage Vb based on the monitoring results of the optical power monitor 52, the first reference light power monitor 53-1, and the second reference light power monitor 53-2.

[0043] As described above, the control circuit that performs MZI control in the optical output circuit 101 includes the first reference light source 35-1 to the Nth reference light source 35-N, the wavelength multiplexer 8, the optical tap circuit 10, the wavelength separation coupler 13, the optical power monitor 52, the first reference light power monitor 53-1 to the Nth reference light power monitor 53-N, and the controller 41.

[0044] Next, it will be explained how the bias condition of the MZI used as the optical filter of the optical output circuit 101 shown in FIG. 1 is monitored. FIG. 2 is a diagram showing the relationship between the optical path difference ΔL of the MZI in the optical output circuit 101 and the optical output intensity of the light of each wavelength output to the output port 7. The optical path difference ΔL is the difference between the optical path length of the first optical waveguide 3 and the optical path length of the second optical waveguide 4. The optical output intensity P(λ) of the light of wavelength λ is monitored by the optical power monitor 52, the optical output intensity P(λ1) of the light of wavelength λ1 is monitored by the first reference light power monitor 53-1, and the optical output intensity P(λ2) of the light of wavelength λ2 is monitored by the second reference light power monitor 53-2.

[0045] First, similar to FIG. 11, consider biasing the MZI to the Null point at wavelength λ using the bias voltage Vb. As shown by the arrow A1 in FIG. 2(a), assume that 0.5λ is selected for the optical path difference ΔL. If the bias voltage Vb is in an appropriate state, as shown by the point A10 in FIG. 2(a), the optical output intensity P(λ) of the light of wavelength λ becomes minimum.

[0046] Here, consider the behavior of the optical output intensity P(λ1) of the first reference light with wavelength λ1 and the optical output intensity P(λ2) of the second reference light with wavelength λ2. In FIG. 2, λ1 = λ / 1.3 and λ2 = λ / 0.7. As shown by points A11 and A12 in FIG. 2(a), when the optical path difference ΔL is 0.5λ, the optical output intensities P(λ1) and P(λ2) are not the minimum and are higher than the optical output intensity P(λ). This state is taken as the reference state.

[0047] FIG. 2(b) shows the optical output intensity P(λ), the optical output intensity P(λ1), and the optical output intensity P(λ2) when bias drift occurs and the optical path difference ΔL increases beyond 0.5λ as indicated by arrow A2. Comparing with the reference state shown in FIG. 2(a), it can be seen that the optical output intensity P(λ) of wavelength λ slightly increases as shown by point A20, the optical output intensity P(λ1) of wavelength λ1 increases rapidly as shown by point A21, and the optical output intensity P(λ2) of wavelength λ2 decreases as shown by point A22.

[0048] FIG. 2(c) shows the optical output intensity P(λ), the optical output intensity P(λ1), and the optical output intensity P(λ2) when bias drift occurs and the optical path difference ΔL decreases beyond 0.5λ as indicated by arrow A3. Comparing with the reference state shown in FIG. 2(a), the optical output intensity P(λ) of wavelength λ slightly increases as shown by point A30, similar to the case of FIG. 2(b). However, different from the case of FIG. 2(b), the optical output intensity P(λ1) of wavelength λ1 decreases as shown by point A31, and the optical output intensity P(λ2) of wavelength λ2 increases rapidly as shown by point A32.

[0049] Based on this information, the controller 41 can determine whether the optical path difference ΔL drifts in the direction of increasing beyond the target value of 0.5λ or in the direction of decreasing from the target value. Then, the controller 41 can feedback the result to the bias voltage Vb. This feedback is realized by the controller 41 increasing or decreasing the bias voltage Vb according to the determination result.

[0050] However, when implementing feedback, the following problems occur. First, when the optical path difference ΔL increases (or decreases) from the optimum value, it is necessary to correctly grasp the correspondence as to whether the bias voltage Vb should be increased or decreased. Since this correspondence is uniquely determined after the manufacture of the MZI, it can be confirmed in the operation test before operation and implemented in the controller 41.

[0051] Another problem is that the behavior of the three types of optical intensities of the optical output intensities P(λ), P(λ1), and P(λ2) changes depending on the selected Null point. The Null point occurs at the optical path difference ΔL that satisfies ΔL = ±λ / 2×(2m + 1), where m is an integer of 0 or more. In FIG. 2, the case where m = 0 and the sign is + was taken as an example for explanation. However, even when m = 0 but the sign is -, that is, when ΔL = -0.5λ, as is clear from FIG. 2, the intensity change of the reference light group at the time of bias drift is opposite to that when ΔL = +0.5λ. Also, when an integer of 1 or more is selected as m, the behavior of the intensity change of each of the optical output intensities P(λ), P(λ1), and P(λ2) is different from that shown at points A11 to A13, points A21 to A23, and points A31 to A33 in FIG. 2.

[0052] To solve this problem, a training period can be provided immediately after starting the control of the MZI. The following operations are performed during the training period.

[0053] First, using the prior art, the controller 41 locks to any Null point that satisfies ΔL = ±λ / 2×(2m + 1) by superimposing (dithering) a pilot tone on the bias voltage Vb and synchronous detection. After the locking is completed, the controller 41 stops the pilot tone and intentionally changes the bias voltage Vb to learn the fluctuations of the three types of optical output intensities P(λ1), P(λ), and P(λ2). When this learning is completed, the training period ends. The training period needs to end in a short time such that the bias drift of the MZI can be ignored.

[0054] After the end of the training period, the controller 41 can monitor and control the bias condition while referring to the learning result. That is, the controller 41 determines whether the optical path difference ΔL is at the optimum value, and if not, determines whether the bias voltage Vb should be increased or decreased, and performs feedback control of the bias voltage Vb.

[0055] In the above description, the case of controlling the MZI to the null point has been described. When controlling the MZI to the quad point, the control may be performed so as to satisfy ΔL = ±λ / 2×(2m + 0.5) or ΔL = ±λ / 2×(2m + 1.5). Here, m is also an integer of 0 or more.

[0056] Also, when controlling the MZI to the peak point, the controller 41 performs control so as to satisfy ΔL = ±mλ. In this case, it is desirable that m is other than 0 and is an integer of 1 or more. This is because, as shown in FIG. 2, in the vicinity of ΔL = 0, the transmission characteristic of the MZI has almost no wavelength dependence, making it difficult to apply this embodiment.

[0057] <Second Embodiment> In the first embodiment, the control circuit for MZI control generates reference lights in parallel and simultaneously, and monitors the input light and the reference light that have passed through the MZI respectively by a plurality of optical power monitors. In the second embodiment, the control circuit of the MZI generates reference lights in time division (time sharing), and monitors the input light and the reference light that have passed through the MZI respectively by one optical power monitor. This embodiment will be described centering on the differences from the first embodiment.

[0058] FIG. 3 is a diagram showing the configuration of the optical output circuit 102 according to the second embodiment. In the optical output circuit 102 shown in FIG. 3, the same parts as those of the optical output circuit 101 according to the first embodiment shown in FIG. 1 are denoted by the same reference numerals, and the description thereof is omitted. Using FIG. 3, the MZI control and its operation of the second embodiment will be described. Also in the optical output circuit 102 shown in FIG. 3, the optical circuit to be controlled by MZI is an optical filter composed of a single MZI.

[0059] The differences between the optical output circuit 102 of the second embodiment shown in FIG. 3 and the optical output circuit 101 of the first embodiment shown in FIG. 1 are that it includes a controller 42 instead of the controller 41, and that it includes only one optical power monitor 11 instead of the (N + 1) types of optical power monitors including the optical power monitor 52 and the first to Nth reference optical power monitors 53-1 to 53-N. The control circuit that performs MZI control in the optical output circuit 102 has a configuration including N reference optical light sources 35, a wavelength division multiplexing coupler 8, an optical tap circuit 10, a wavelength demultiplexing coupler 13, an optical power monitor 11, and a controller 42. In this embodiment as well, the case where N = 2 will be described as an example.

[0060] The controller 42 controls the first reference optical light source 35-1 and the second reference optical light source 35-2 to be in an extinction or emission state respectively. When both the first reference optical light source 35-1 and the second reference optical light source 35-2 are in the extinction state, the monitor result of the optical power monitor 11 is the optical output intensity P(λ) at wavelength λ.

[0061] On the other hand, when the controller 42 controls the first reference optical light source 35-1 to emit light and the second reference optical light source 35-2 to be in the extinction state, the monitor result of the optical power monitor 11 is determined by both the optical output intensity P(λ) at wavelength λ and the optical output intensity P(λ1) at wavelength λ1. This monitor result is denoted as the optical output intensity P(λ, λ1).

[0062] When the wavelength dependence of the sensitivity of the optical power monitor 11 can be ignored, the optical output intensity P(λ1) at wavelength λ1 can be obtained simply as the difference obtained by subtracting the optical output intensity P(λ) from the optical output intensity P(λ, λ1). When the wavelength dependence of the optical power monitor 11 cannot be ignored, the difference between the two may be calculated after multiplying by a correction coefficient. These operations can be easily realized by performing analog / digital conversion of the monitor result inside the controller 42 and performing the operations digitally.

[0063] When the controller 42 controls so that the first reference light source 35-1 is extinguished and the second reference light source 35-2 emits light, the monitoring result of the optical power monitor 11 is determined by both the optical output intensity P(λ) of the light with wavelength λ and the optical output intensity P(λ2) of the light with wavelength λ2. By the same procedure as in the case of the first reference light above, the optical output intensity P(λ2) of wavelength λ2 can be obtained.

[0064] After the controller 42 obtains the information on these three types of optical output intensities, i.e., the optical output intensity P(λ), the optical output intensity P(λ1), and the optical output intensity P(λ2), it can control the bias voltage Vb in the same manner as in the first embodiment.

[0065] As described above, the controller 42 controls so that all of the N reference light sources 35 are extinguished or only one of them emits light. When all of the N reference light sources 35 are extinguished, the input light with wavelength λ that has passed through the optical filter is output to the output port 7. When the n-th reference light source 35-n emits light and the other reference light sources 35 are extinguished, the input light with wavelength λ that has passed through the optical filter and the light multiplexed with the n-th reference light with wavelength λ n are output. The optical power monitor 11 monitors the optical output intensity P(λ) of the light with wavelength λ when all of the N reference light sources 35 are extinguished, and obtains the optical output intensity P(λ, λ n ) when only the n-th reference light source 35-n emits light. The controller 42 subtracts the optical output intensity P(λ) from the optical output intensity P(λ, λ n ) to obtain the optical output intensity P(λ n ) of the n-th reference light. The controller 42 controls the bias voltage Vb in the same manner as in the first embodiment using the optical output intensity P(λ) and the optical output intensities P(λ1) to P(λ N ).

[0066] The second embodiment is more complicated in terms of requiring calculations compared to the first embodiment. However, it has the advantage that only one optical power monitor is needed.

[0067] <The Third Embodiment> In the second embodiment, the control circuit of the MZI generated reference lights of a plurality of wavelengths by respective ones of a plurality of reference light sources. In this embodiment, the control circuit of the MZI generates reference lights of a plurality of wavelengths in a time-division manner by a single reference light source. Further, the control circuit of the MZI in this embodiment generates a reference light of an optical pulse train, and obtains the optical intensity of the reference light that has passed through the MZI by synchronous detection. This embodiment will be described centering on the differences from the above-described embodiments.

[0068] FIG. 4 is a diagram showing a configuration of an optical output circuit 103 according to a third embodiment. In the optical output circuit 103 shown in FIG. 4, the same parts as those of the optical output circuit 101 according to the first embodiment shown in FIG. 1 and the optical output circuit 102 according to the second embodiment shown in FIG. 3 are denoted by the same reference numerals, and the description thereof is omitted. Using FIG. 4, the MZI control and its operation according to the third embodiment will be described. Also in the optical output circuit 103 shown in FIG. 4, the optical circuit to be controlled by the MZI control is an optical filter composed of a single MZI.

[0069] The configuration of the optical output circuit 103 according to the third embodiment shown in FIG. 4 is different from the configuration of the optical output circuit 102 according to the second embodiment shown in FIG. 3 in that an optical coupler 33 is provided instead of the wavelength division multiplexer 8, a wavelength tunable light source 32 is provided instead of the N reference light sources 35, a controller 43 is provided instead of the controller 42, and a synchronous detection / averaging processing circuit 34 is further provided. The control circuit that performs MZI control in the optical output circuit 103 has a configuration including a wavelength tunable light source 32, an optical coupler 33, an optical tap circuit 10, an optical power monitor 11, a synchronous detection / averaging processing circuit 34, and a controller 42.

[0070] The wavelength tunable light source 32 generates a first reference light to an Nth reference light. Also in this embodiment, the case where N = 2 will be described as an example. The optical coupler 33 outputs light in which the input light of wavelength λ is multiplexed with the first reference light of wavelength λ1 or the second reference light of wavelength λ2 generated by the wavelength tunable light source 32. The optical branching circuit 2 branches the light multiplexed by the optical coupler 33 into two. Further, the synchronous detection / averaging processing circuit 34 performs synchronous detection or averaging processing on the monitoring result by the optical power monitor 11.

[0071] The wavelength-variable light source 32 is controlled by a wavelength control signal output from the controller 43. The wavelength control signal has three values. The wavelength-variable light source 32 takes one of the states of generating an optical pulse train with wavelength λ1, generating an optical pulse train with wavelength λ2, or extinguishing, according to the value of the wavelength control signal. The pulse interval and phase of the optical pulse train are determined by a reference clock output from the controller 43.

[0072] First, the case where the output of the wavelength-variable light source 32 is extinguished by the wavelength control signal output from the controller 43 will be described. At this time, the monitoring result output from the optical power monitor 11 is the optical output intensity P(λ).

[0073] The controller 43 also outputs the wavelength control signal to the synchronous detection / averaging processing circuit 34. When the extinction of the wavelength-variable light source 32 is instructed by the wavelength control signal, the synchronous detection / averaging processing circuit 34 performs an averaging process on the monitoring result output from the optical power monitor 11. Thereby, noise such as dark current superimposed on the optical power monitor 11 is suppressed.

[0074] Next, the case where the wavelength-variable light source 32 generates an optical pulse train with wavelength λ1 by the wavelength control signal output from the controller 43 will be described. At this time, the monitoring result output from the optical power monitor 11 periodically repeats the optical output intensity P(λ) and the optical output intensity P(λ, λ1).

[0075] The controller 43 also outputs a wavelength control signal to the synchronous detection / averaging processing circuit 34. When the wavelength variable light source 32 is instructed to generate an optical pulse train with a wavelength of λ1, the synchronous detection / averaging processing circuit 34 performs synchronous detection on the monitoring result output by the optical power monitor 11. The controller 43 outputs a reference clock to the synchronous detection / averaging processing circuit 34 in addition to the wavelength variable light source 32. The synchronous detection / averaging processing circuit 34 performs synchronous detection with reference to this reference clock. As a result, the synchronous detection / averaging processing circuit 34 obtains an amount corresponding to the difference between the optical output intensity P(λ, λ1) and the optical output intensity P(λ), that is, the optical output intensity P(λ1) as the synchronous detection result. When the wavelength dependence of the optical power monitor 11 cannot be ignored or losses due to filtering during synchronous detection cannot be ignored, a correction coefficient may be added to the synchronous detection result inside the controller 43.

[0076] When the wavelength variable light source 32 generates an optical pulse train with a wavelength of λ2 by the wavelength control signal output by the controller 43, the monitoring result output by the optical power monitor 11 periodically repeats the optical output intensity P(λ) and the optical output intensity P(λ, λ2). The controller 43 obtains the optical output intensity P(λ2) by the same synchronous detection process as in the case of the first reference light described above.

[0077] After the controller 43 obtains information on three types of optical intensities, namely the optical output intensity P(λ), the optical output intensity P(λ1), and the optical output intensity P(λ2), it controls the bias voltage Vb in the same manner as in the first embodiment.

[0078] As described above, the controller 43 uses the wavelength control signal to instruct whether the wavelength variable light source 32 generates an optical pulse train with a wavelength of λ1 to λ N of the optical pulse train or extinguishes it. When the output of the wavelength variable light source 32 is extinguished, the synchronous detection / averaging processing circuit 34 performs averaging processing on the monitoring result output by the optical power monitor 11 to obtain the optical output intensity P(λ). When the wavelength variable light source 32 has a wavelength of λ based on the reference clock nWhen generating an optical pulse train, the synchronous detection / averaging processing circuit 34 performs synchronous detection of the monitor result output by the optical power monitor 11 with reference to the reference clock. The synchronous detection / averaging processing circuit 34 obtains the optical output intensity P(λ n ), which is the difference between the optical output intensity P(λ, λ n ) and the optical output intensity P(λ). The controller 43 controls the bias voltage Vb in the same manner as in the first embodiment using the optical output intensity P, the optical output intensities P(λ1) to P(λ N ).

[0079] In this embodiment, although synchronous detection is used, no dithering is applied to the input light of wavelength λ and the bias voltage Vb. Therefore, no noise is superimposed on the transmitted light. Also, it has the advantage that noise derived from circuit noise of the optical power monitor and power fluctuations of the light source can be suppressed.

[0080] <Variation of the Third Embodiment> In the third embodiment, the optical output intensity P, the optical output intensities P(λ1) and P(λ2) are measured, and the drift of the bias is monitored using these measured values. However, in order to measure P(λ) with a single optical power monitor 11, in order to suppress the circuit noise of the optical power monitor 11, it is necessary to secure a certain length of time for extinguishing the reference light and perform processing such as averaging the output of the optical power monitor 11. The longer the averaging processing time, the better the noise is suppressed, but on the other hand, there is a problem that the processing speed becomes slow.

[0081] Here, we again focus on FIG. 2. If the values of the optical output intensities P(λ1) and P(λ2) in the vicinity of the optimal bias point, that is, the values of points A11, A12, A21, A22, A31, and A32, are well understood, it is possible to estimate the bias drift without measuring the value of the optical output intensity P(λ), that is, points A10, A20, and A30. This is because, as described above with reference to FIGS. 2(b) and 2(c), when the optical path difference ΔL increases from the optimal value, the optical output intensity P(λ1) increases rapidly and the optical output intensity P(λ2) decreases, while when the optical path difference ΔL decreases from the optimal value, the optical output intensity P(λ1) decreases and the optical output intensity P(λ2) increases rapidly. Therefore, during the training period provided when the optical output circuit 103 is started up, the controller 43 learns the changes in the optical output intensities P(λ1) and P(λ2) in the vicinity of the optimal bias point, and records the learned changes as a database, so that it is also possible to perform bias control without measuring the optical output intensity P(λ). That is, the controller 43 estimates the bias drift by comparing the changes in the optical output intensities P(λ1) and P(λ2) recorded in the database with the measured changes in the optical output intensities P(λ1) and P(λ2), and corrects the optical path difference by controlling the bias voltage Vb based on the estimated bias drift.

[0082] FIG. 5 is a diagram schematically showing the temporal changes in the outputs of the wavelength-variable light source 32 and the optical power monitor 11 in such an embodiment. FIG. 5(a) shows the temporal change of the wavelength control signal, FIG. 5(b) shows the temporal change of the output of the wavelength-variable light source 32, FIG. 5(c) shows the temporal change of the reference clock, and FIG. 5(d) shows the temporal change of the output of the optical power monitor 11.

[0083] As shown in FIG. 5(a), the wavelength control signal has three values: a value indicating the generation of an optical pulse train with wavelength λ1, a value indicating the generation of an optical pulse train with wavelength λ2, and a value indicating extinction, but the time period corresponding to the extinction period is made short. The extinction period is the period during which the value indicating extinction is output.

[0084] As shown in Fig. 5(b), the wavelength-variable light source 32 assumes one of the following states: generating an optical pulse train with wavelength λ1, generating an optical pulse train with wavelength λ2, or having a short extinction time, depending on the value of the control signal. In Fig. 5(b), the wavelength-variable light source 32 generates a pulse train with wavelength λ1 during the period from time 0 to 1, wavelength λ2 during the period from time 1 to 2, and wavelength λ1 again during the period from time 2 to 3, but provides a short extinction period when changing the wavelength. This is because most wavelength-variable light sources require a certain amount of time to change the wavelength. When the time required for the wavelength change of the wavelength-variable light source is sufficiently short, the extinction period can be omitted and the wavelength control signal can be made binary.

[0085] In the time period when the wavelength of the output from the wavelength-variable light source 32 is the same, the period of the optical pulse is T. This period T and the phase of the optical pulse are determined by the reference clock. Here, it is assumed that the optical pulse is generated when the reference clock is at a high level, as shown in Fig. 5(c).

[0086] Light of wavelength λ is always input to the optical power monitor 11 regardless of the value of the wavelength control signal. Therefore, as shown in Fig. 5(d), the minimum level of the output of the optical power monitor 11 is always maintained at the optical output intensity P(λ). However, by the above-described procedure, the synchronous detection / averaging processing circuit 34 performs synchronous detection processing with reference to the reference clock to obtain the values of the optical output intensity P(λ1) corresponding to the difference between the optical output intensity P(λ, λ1) and the optical output intensity P(λ), and the optical output intensity P(λ2) corresponding to the difference between the optical output intensity P(λ, λ2) and the optical output intensity P(λ).

[0087] The circuit noise of the optical power monitor 11 is also input to the synchronous detection / averaging processing circuit 34. However, frequency components other than 1 / T are suppressed during the process of performing synchronous detection. Therefore, the circuit noise of the optical power monitor 11 is suppressed.

[0088] In the above description, the reference light with wavelength λ1 and the reference light with wavelength λ2 are generated using only the wavelength-variable light source 32. However, a plurality of reference light sources 35 may be used as in the second embodiment shown in FIG. 3. That is, these plurality of reference light sources 35 may generate optical pulse trains in a time-division manner, and a synchronous detection / averaging processing circuit 34 may perform synchronous detection processing. In this case, the optical output circuit 103 includes N reference light sources 35 and a wavelength division multiplexer 8 instead of the wavelength-variable light source 32 and the optical coupler 33. Further, the controller 41 of the optical output circuit 101 in the first embodiment may control the bias voltage Vb in the same manner as the controller 43 in this embodiment based on the optical output intensities P(λ1) and P(λ2) output from the first reference light power monitor 53-1 and the second reference light power monitor 53-2, respectively.

[0089] Also, in the above description, the synchronous detection / averaging processing circuit 34 and the controller 43 are separate circuits, and the synchronous detection / averaging processing circuit 34 receives a multi-valued wavelength control signal issued from the controller 43. However, when the synchronous detection circuit is configured digitally, the synchronous detection / averaging processing circuit 34 and the controller 43 may be integrated, and the generation of the optical pulse train, wavelength change, start / stop of the synchronous detection processing may be performed according to a pre-programmed flowchart. When such a configuration is adopted, the wavelength control signal can be omitted.

[0090] <Fourth Embodiment> In the fourth embodiment, the reference light is transmitted through the optical circuit in the direction opposite to the input light. This embodiment will be described centering on the differences from the above-described embodiments. Hereinafter, this embodiment will be described centering on the differences from the first embodiment, but the same differences can be applied to the other above-described embodiments.

[0091] FIG. 6 is a diagram showing the configuration of the optical output circuit 104 according to the fourth embodiment. In the optical output circuit 104 shown in FIG. 6, the same parts as those of the optical output circuit 101 according to the first embodiment shown in FIG. 1 are denoted by the same reference numerals, and the description thereof is omitted. Using FIG. 6, the MZI control and its operation according to the fourth embodiment will be described. Also in the optical output circuit 104 shown in FIG. 6, the optical circuit to be MZI-controlled is an optical filter composed of a single MZI.

[0092] The differences between the configuration of the optical output circuit 104 of the fourth embodiment shown in FIG. 6 and the configuration of the optical output circuit 101 of the first embodiment shown in FIG. 1 are that a controller 44 is provided instead of the controller 41, a wavelength demultiplexer 61 is provided instead of the wavelength multiplexer 8, a wavelength multiplexer 62 is provided instead of the wavelength demultiplexer 13, N reference light sources 35 are connected to the wavelength multiplexer 62, N reference light power monitors 53 are connected to the wavelength demultiplexer 61, and the optical bandpass filter 12 is not provided. Also in this embodiment, the case of N = 2 will be described as an example.

[0093] The wavelength demultiplexer 61 outputs the input light of wavelength λ, and the optical branching circuit 2 branches the input light of wavelength λ output by the wavelength demultiplexer 61 into two. The optical multiplexing circuit 6 multiplexes the light propagated through the first optical waveguide 3 and the light propagated through the second optical waveguide 4, and outputs the multiplexed light of wavelength λ to the output port 7. The optical tap circuit 10 taps the output light of the optical multiplexing circuit 6 to the optical filter. The wavelength multiplexer 62 outputs the input light of wavelength λ tapped by the optical tap circuit 10 to the optical power monitor 52.

[0094] On the other hand, the wavelength division multiplexing coupler 62 inputs the first reference light output from the first reference light source 35-1 and the second reference light output from the second reference light source 35-2 into the optical filter in the direction opposite to the input light. The wavelength separation coupler 61 inputs the first reference light and the second reference light that have passed through the MZI and are output to the optical input port 1. The wavelength separation coupler 61 outputs the first reference light with wavelength λ1 to the first reference light power monitor 53-1, and outputs the second reference light with wavelength λ2 to the second reference light power monitor 53-2. The controller 44 controls the bias voltage Vb in the same manner as in the first embodiment based on the monitoring results of the optical power monitor 52, the first reference light power monitor 53-1, and the second reference light power monitor 53-2.

[0095] As described above, the optical power monitor 52 monitors the input light that has passed through the optical filter to obtain the optical output intensity P(λ). The wavelength division multiplexing coupler 62 multiplexes the first to Nth reference lights and inputs them into the MZI in the direction opposite to the input light. The nth reference light power monitor 53-n monitors the nth reference light with wavelength λ that has passed through the optical filter and is separated by the wavelength separation coupler 61 to obtain the optical output intensity P(λ n ) n . The controller 44 controls the bias voltage Vb in the same manner as in the first embodiment using the optical output intensities P(λ), P(λ1),..., P(λ N ).

[0096] The optical filter constituted by the MZI has the same transmission characteristics even when the input and output are interchanged. Therefore, even if the arrangement like the optical output circuit 104 is adopted, the bias voltage Vb can be controlled by the same procedure as in the first embodiment. Further, compared with the first embodiment, there is an advantage that the optical bandpass filter 12 can be omitted.

[0097] <Variations of the First to Fourth Embodiments> In the first to fourth embodiments, all the optical filters consisted of a single MZI. However, as described above, by disposing modulation electrodes on either one or both of the first optical waveguide 3 and the second optical waveguide 4 to apply a modulation signal and using a non-linear optical effect such as the Pockels effect, an optical modulator consisting of a single MZI can be configured. Even in this case, control of the bias voltage Vb is necessary, and it is necessary to bias the MZI to Null or Quad etc. according to the signal format. However, even in this case, the bias voltage Vb can be adjusted by the same method as in the first to fourth embodiments.

[0098] However, when the MZI is used as a modulator, it is necessary to note that the output characteristics of the MZI are different from the output characteristics shown in FIG. 2. In FIG. 2, the optical output intensity is extinguished when the optical path difference ΔL is at the Null point and becomes the maximum intensity when it is at the Peak point. However, when modulation is applied to the MZI, the optical phase inside the MZI always fluctuates around the operating point. Therefore, even when the optical path difference ΔL is at the Null point or the Peak point, the optical output does not completely extinguish or reach the maximum intensity. However, the optical output intensity becomes the minimum or maximum within the possible range. Also, depending on the non-linear optical effect used for modulation, the modulation efficiency strongly depends on the wavelength and the propagation direction of light. Therefore, the light intensity of the reference light with different wavelengths, especially the reference light traveling in the reverse direction as in the fourth embodiment, becomes more complicated than the transmission characteristics shown in FIG. 2.

[0099] However, even in such a case, it is certain that the intensity of the optical output is determined by the optical path difference ΔL and the wavelength. Therefore, during the training period described in the first embodiment, the controllers 41 to 44 measure the changes in various optical output intensities P(λ) and P(λ1) to P(λ N ) accompanying the fluctuation of the optical path difference ΔL. For example, when N = 2, by learning the changes in various optical output intensities of P(λ1), P(λ), and P(λ2), it becomes possible to control the bias voltage Vb.

[0100] Also, in the first to fourth embodiments, the optical circuit to be controlled by the MZI had a single MZI, but it may have a plurality of MZIs. The controller controls at least one of the optical path differences between the two optical paths of each of the plurality of MZIs. It is desirable that the intervals between the wavelengths of the N types of reference lights and the wavelength of the input light do not match any of the free spectral ranges of the MZI group of the optical circuit.

[0101] <Fifth Embodiment> In the above-described embodiments, the optical circuit to be controlled by the MZI had a single MZI. The optical circuit to be controlled by the MZI in this embodiment has a plurality of nested MZIs.

[0102] FIG. 7 is a diagram showing the configuration of the optical output circuit 105 according to the fifth embodiment. Using FIG. 7, the MZI control and its operation in the fifth embodiment will be described. In the optical output circuit 105 shown in FIG. 7, the optical circuit to be controlled by the MZI is not a single MZI but a nested MZI group that constitutes an IQ optical modulator. The optical output circuit 105 of the fifth embodiment has a configuration in which the single MZI in the optical output circuit 103 of the third embodiment shown in FIG. 4 is replaced with the IQ optical modulator 92 shown in FIG. 12.

[0103] That is, the difference between the configuration of the optical output circuit 105 shown in FIG. 7 and the configuration of the IQ optical modulator 92 shown in FIG. 12 is that it includes a control circuit having a wavelength-variable light source 32, an optical coupler 33, an optical power monitor 11, an optical band-pass filter 12, a synchronous detection / averaging processing circuit 34, and a controller 45. The wavelength-variable light source 32, the optical coupler 33, the optical power monitor 11, the optical band-pass filter 12, and the synchronous detection / averaging processing circuit 34 operate in the same manner as in the third embodiment.

[0104] To generate a QAM signal, as described above, the optical path difference of the In-Phase MZI 50 is biased by the bias voltage Vb_I applied via the I-side bias application electrode 23 so as to extinguish at the moment when the modulation signal I is zero. Also, the optical path difference of the Quadrature MZI 51 is biased by the bias voltage Vb_Q applied via the Q-side bias application electrode 27 so as to extinguish at the moment when the modulation signal Q is zero.

[0105] Here, the optical path difference of the In-Phase MZI 50 at the moment when the modulation signal I and the modulation signal Q are zero is defined as ΔL_I, and the optical path difference of the Quadrature MZI 51 is defined as ΔL_Q.

[0106] m I and m Q When m and m are integers of 0 or more, in order to satisfy the above conditions, the following equations (1) and (2) must be satisfied.

[0107] ΔL_I = ±λ / 2 × (2m I +1) …(1)

[0108] ΔL_Q = ±λ / 2 × (2m Q +1) …(2)

[0109] In this embodiment, during the aforementioned training period, m I and m Q are set to different values.

[0110] In the following simulation, m I = 0, m Q = 1 are set, and it is assumed that ΔL_I = -0.5λ and ΔL_Q = 1.5λ are selected during the training period. Also, the optical phase difference between the optical output of the In-Phase MZI 50 and the optical output of the Quadrature MZI 51 must also be set to ±π / 4 + π × m ph during the aforementioned training period. In the following simulation, m phLet it be assumed that =0 is set and the optical path difference ΔL_Ph = -0.25λ of the parent MZI of the optical output circuit 105 is selected during the training period.

[0111] FIG. 8 is a diagram showing the results of simulating the variations in the respective optical output intensities P(λ2), P(λ), P(λ1) of the input light at the time of bias drift and the two reference lights, the first reference light and the second reference light. Here, λ2 is 1563 nm, λ is 1546 nm, and λ1 is 1529 nm. Also, it is assumed that the QAM signal is 16-valued.

[0112] FIG. 8(a) shows the changes when a bias drift occurs in the optical path difference ΔL_I adjusted by the bias voltage Vb_I. The horizontal axis shows the change in the optical path difference ΔL_I. The center of the horizontal axis is the set target ΔL_I = -0.5λ. The vertical axis shows the optical output intensity. The graph on the right side of FIG. 8(a) is an enlarged view near the set target of the graph on the left side of FIG. 8(a). Different from FIG. 2, since the light is modulated randomly, even when the optical path difference ΔL_I is optimal, complete extinction is not achieved. Also, since there is a wavelength dependence in the modulation efficiency, the maximum optical intensities of P(λ2), P(λ), P(λ1) do not match.

[0113] FIG. 8(b) shows the changes when a bias drift occurs in the optical path difference ΔL_Q adjusted by the bias voltage Vb_Q. The horizontal axis shows the change in the optical path difference ΔL_Q. The center of the horizontal axis is the set target ΔL_Q = 1.5λ. The vertical axis shows the optical output intensity. The graph on the right side of FIG. 8(b) is an enlarged view near the set target of the graph on the left side of FIG. 8(b).

[0114] FIG. 8(c) shows the changes when a bias drift occurs in the optical path difference ΔL_Ph adjusted by the bias voltage Vb. The horizontal axis shows the change in the optical path difference ΔL_Ph. The center of the horizontal axis is the set target ΔL_Ph = -0.25λ. The vertical axis shows the optical output intensity. Note that the scale of the vertical axis is different from that of FIGS. 8(a) and 8(b).

[0115] Based on the relative relationships of the three types of optical intensities, namely the optical output intensities P(λ2), P(λ), and P(λ1), it is possible to determine which bias has drifted. Figure 9 is a diagram for more clearly explaining this. The simulation parameters and each horizontal axis are the same as those in Figure 8. However, in Figure 9, what is plotted is not the three types of optical intensities but the differences between two types of optical output intensities, P(λ2) - P(λ) and P(λ1) - P(λ). In Figure 9, these differences in optical output intensities are shown as functions of the optical path differences ΔL_I, ΔL_Q, and ΔL_Ph respectively.

[0116] Looking at the slope near the center of the horizontal axis in Figure 9(a), when the optical path difference ΔL_I adjusted by the bias voltage Vb_I starts to increase due to drift from -0.5λ, the optimal value set as the target, P(λ2) - P(λ) increases, while P(λ1) - P(λ) decreases.

[0117] Looking at the slope near the center of the horizontal axis in Figure 9(b), when the optical path difference ΔL_Q adjusted by the bias voltage Vb_Q starts to increase due to drift from 1.5λ, the optimal value set as the target, P(λ2) - P(λ) decreases, while P(λ1) - P(λ) increases.

[0118] Looking at the slope near the center of the horizontal axis in Figure 9(c), when the optical path difference ΔL_Ph adjusted by the bias voltage Vb starts to increase due to drift from -0.25λ, the optimal value set as the target, both P(λ2) - P(λ) and P(λ1) - P(λ) decrease.

[0119] The controller 45 can determine these pieces of information and decide which bias should be corrected. However, this determination is not unrestrictedly possible. If a drift of a certain magnitude or more occurs, it will lead to an incorrect determination. In FIG. 9(a), when the optical path difference ΔL_I drifts from the optimal value of -0.5λ to -0.45λ, the slopes of P(λ2)-P(λ) and P(λ1)-P(λ) reverse. Therefore, the error allowed at the end of the training period should be made sufficiently smaller than 0.05λ (equivalent to 0.1π in optical phase). Furthermore, when a drift occurs, it is necessary to quickly correct it to the optimal value before the magnitude of the drift reaches 0.05λ.

[0120] <Sixth Embodiment> In the fifth embodiment, when ΔL_I = ±λ / 2×(2m I +1) and ΔL_Q = ±λ / 2×(2m Q +1), in the aforementioned training period, m I and m Q were set to be different values. However, there are limitations in the dynamic range of the voltage that can be applied to the bias voltage Vb_I and the bias voltage Vb_Q due to the device configuration. Therefore, there may be cases where it is technically impossible to set m I and m Q to different values.

[0121] m I and m QWhen they become the same value, there is no essential difference between the behaviors of the in-phase MZI 50 and the quadrature MZI 51. Therefore, it becomes impossible to discriminate between the drift of the optical path difference ΔL_I and the drift of the optical path difference ΔL_Q. This is because the changes in the optical output intensities P(λ1), P(λ), and P(λ2) when ΔL_I = ΔL_Q are the same whether the optical path difference ΔL_I is increasing or the optical path difference ΔL_Q is increasing, and are also the same whether the optical path difference ΔL_I is decreasing or the optical path difference ΔL_Q is decreasing. Alternatively, the changes in P(λ1), P(λ), and P(λ2) when ΔL_I = -ΔL_Q are the same whether the optical path difference ΔL_I is increasing or the optical path difference ΔL_Q is decreasing, and are also the same whether the optical path difference ΔL_I is decreasing or the optical path difference ΔL_Q is increasing.

[0122] To solve this problem, the following control procedure is followed. In the following description, the drive amplitude of the modulator is assumed to be small enough that the non-linearity of the modulator can be ignored. First, when the controller 45 determines that either the optical path difference ΔL_I or the optical path difference ΔL_Q is increasing, it first slightly changes the bias voltage Vb_I so as to decrease ΔL_I. Alternatively, when the controller 45 determines that either the optical path difference ΔL_I or the optical path difference ΔL_Q is decreasing, it first slightly changes the bias voltage Vb_I so as to increase ΔL_I. As a result, if the controller 45 determines that the optical output intensity P(λ) is decreasing, it can be determined that the drifted one is the optical path difference ΔL_I.

[0123] Conversely, if the light output intensity P(λ) appears to increase as a result of a slight change in the bias voltage Vb_I described above, the controller 45 can determine that it is the optical path difference ΔL_Q that has drifted. Therefore, the controller 45 quickly returns the bias voltage Vb_I to its original value. After that, the controller 45 corrects the optical path difference ΔL_I or the optical path difference ΔL_Q that has drifted until it returns to the values of the light output intensities P(λ1), P(λ), and P(λ2) learned at the end of the training period. As a result, the controller 45 of the optical output circuit 105 controls to correct the optical path difference ΔL_I or the optical path difference ΔL_Q so that the light output intensity P(λ) decreases. In the description so far, the drive amplitude of the modulator has been assumed to be small enough that the non-linearity of the modulator can be ignored. However, in the case of a relatively simple signal format such as a 4-value QAM signal, since the penalty due to the non-linearity of the modulator is small, there may be an operation in which the drive amplitude of the modulator is increased. In such a case, the light output intensity P(λ) becomes large when the above bias voltage Vb_I or Vb_Q is optimal. Therefore, when using such a signal format, the controller 45 of the optical output circuit 105 controls to correct the optical path difference ΔL_I or the optical path difference ΔL_Q so that the light output intensity P(λ) increases.

[0124] According to this embodiment, it is possible to optimally adjust the bias of an optical filter or an optical modulator composed of one or more MZIs without applying periodic minute fluctuations by a pilot tone (dithering) to one or more bias voltages. For this reason, it is possible to suppress the optical noise superimposed on the optical signal due to minute fluctuations in the bias voltage.

[0125] According to the above-described embodiment, the control circuit controls at least one of the optical path differences of the two optical paths included in each of the single or multiple MZIs included in the optical circuit. The reference light generation unit generates N types (N is a natural number) of wavelengths λ1 to λ that are different from the wavelength λ of the input light input to the optical circuit. NEach reference light is generated in parallel or in a time-division manner. The reference light generation unit corresponds to, for example, the first reference light source 35-1 to the Nth reference light source 35-N and the wavelength-variable light source 32 in the embodiment. The reference light input unit inputs the reference light into the optical circuit. The reference light input unit corresponds to, for example, the wavelength-division multiplexing coupler 8, the optical coupler 33, and the wavelength-division multiplexing coupler 62 in the embodiment. The optical power monitor unit monitors the optical intensity of the propagated input light, which is the input light after propagating through a single or a plurality of MZIs in the optical circuit, the optical intensity of the propagated reference light, which is the reference light after propagating through the single or a plurality of MZIs that the propagated input light has passed through, or the optical intensity of the light obtained by multiplexing the propagated input light and the propagated reference light. The optical power monitor unit corresponds to, for example, the optical power monitor 52, the Nth reference optical power monitor to the Nth reference optical power monitor 53-N, and the optical power monitor 11 in the embodiment. The controller controls to correct the optical path difference based on the optical intensity P(λ) of the propagated input light at wavelength λ obtained using the monitoring result by the optical power monitor unit and the optical intensities P(λ1) to P(λ N of each of the N types of propagated reference lights P(λ1) to P(λ N ). That is, the controller determines whether the optical path difference between the two optical paths of each of the single or a plurality of MZIs is at the target value by mutually comparing the optical intensity P(λ) of the propagated input light and the optical intensities P(λ1) to P(λ N ) of each of the N types of propagated reference lights, and corrects it if it is not at the target value. Alternatively, the controller controls to correct the optical path difference based on the optical intensities P(λ1) to P(λ N of each of the N types of propagated reference lights P(λ1) to P(λ N ) obtained using the monitoring result by the optical power monitor unit.

[0126] The reference light input unit may input the N types of reference lights into the optical circuit by multiplexing them with the input light. In this case, the control circuit may further include a wavelength separation unit. The wavelength separation unit wavelength-separates the propagated input light and the N types of propagated reference lights output by multiplexing from the optical circuit. The wavelength separation unit is, for example, the wavelength separation coupler 13 in the embodiment. The optical power monitor unit detects the optical intensity of each of the propagated input light and the N types of propagated reference lights separated by the wavelength separation unit.

[0127] The controller may control the extinction of the reference light and the generation of the reference light by time division in the reference light generation unit. The reference light input unit multiplexes the reference light generated by time division with the input light and inputs the multiplexed light into the optical circuit. The optical power monitor unit monitors the optical intensity P(λ) of the input light after propagation at wavelength λ, and the optical intensity P(λ, λ n (where n is a natural number from 1 to N) of the light obtained by monitoring the light in which the propagated reference light is multiplexed, and outputs the optical intensity P(λ, λ n ) to the controller. The controller calculates the optical intensities P(λ1) to P(λ n ) of the optical intensity reference lights at wavelengths λ1 to λ N by comparing and calculating the optical intensity P(λ) and the optical intensity P(λ, λ N ).

[0128] The controller may control the generation of the reference light by time division in the reference light generation unit. The reference light input unit multiplexes the reference light with the input light and inputs the multiplexed light into the optical circuit. The optical power monitor unit monitors the optical intensity P(λ, λ n (where n is a natural number from 1 to N) of the light obtained by monitoring the light in which the propagated reference light is multiplexed, and outputs the optical intensity P(λ, λ n ) to the controller. The controller calculates the optical intensities P(λ1) to P(λ n ) of the propagated reference lights at wavelengths λ1 to λ N based on the optical intensity P(λ, λ N ).

[0129] The controller may control the extinction of the reference light and the generation of the optical pulse train of the reference light by a pulse train in the reference light generation unit. The reference light input unit multiplexes the optical pulse train of the reference light generated by time division with the input light and inputs the multiplexed light into the optical circuit. The control circuit further includes a synchronous detection unit. The synchronous detection unit corresponds to the synchronous detection / averaging processing circuit 34 in the embodiment. The synchronous detection unit averages the result of monitoring the light output from the optical circuit while the reference light generation unit is in the extinction state in the optical power monitor unit to obtain the optical intensity P(λ) of the input light after propagation at wavelength λ, and when the reference light generation unit is at wavelength λ nWhile generating the optical pulse train of the reference light (where n is a natural number from 1 to N), the result of monitoring the light output from the optical circuit in the optical power monitor unit is synchronously detected by the reference clock used for generating the optical pulse train, and the optical intensity P(λ n ) of the reference light after propagation at wavelength λ n ) is obtained.

[0130] When the reference light generated in the reference light generation unit is an optical pulse train, the controller may control the extinction of the reference light and the generation of the optical pulse train in the reference light generation unit. The period and phase of the optical pulse train of the reference light are determined by the reference clock generated by the controller. The reference light input unit multiplexes the optical pulse train of the reference light generated in a time-division manner with the input light and inputs it to the optical circuit. The control circuit further includes a synchronous detection unit. The synchronous detection unit averages the result of monitoring the light output from the optical circuit in the optical power monitor unit while the reference light generation unit is in extinction, and obtains the optical intensity P(λ) of the input light after propagation at wavelength λ. When the reference light generation unit generates the optical pulse train of the reference light at wavelength λ n (where n is a natural number from 1 to N), the result of monitoring the light output from the optical circuit in the optical power monitor unit is synchronously detected by the reference clock used for generating the optical pulse train, and the optical intensity P(λ n ) of the reference light after propagation at wavelength λ n ) is obtained.

[0131] When the reference light generated in the reference light generation unit is an optical pulse train, the controller may control the generation of the optical pulse train of the reference light in the reference light generation unit. The period and phase of the optical pulse train of the reference light are determined by the reference clock generated by the controller. The reference light input unit multiplexes the optical pulse train of the reference light generated in a time-division manner with the input light and inputs it to the optical circuit. The control circuit further includes a synchronous detection unit. The synchronous detection unit synchronously detects the result of monitoring the light output from the optical circuit in the optical power monitor unit while the reference light generation unit generates the optical pulse train of the reference light at wavelength λ n , and obtains the optical intensity P(λ n ) of the reference light after propagation at wavelength λ (where n is a natural number from 1 to N). n ) is obtained.

[0132] The reference light may propagate through one or more MZIs through which the input light has propagated in a direction opposite to that of the input light.

[0133] The intervals between the wavelengths of the N types of reference lights and the wavelength of the input light may each not match any of the free spectral ranges of one or more MZIs of the optical circuit.

[0134] In any one of the one or more MZIs of the optical circuit, when the optical path difference is set so that the interference intensity of the input light with wavelength λ is maximized, the optical path difference may be k times the wavelength λ (k is a non-zero integer).

[0135] The optical circuit may be an optical modulator. In this case, a modulation signal is applied to at least one of the one or more MZIs of the optical circuit.

[0136] The optical circuit may be an IQ optical modulator having an nested MZI in which an In-Phase MZI is arranged in one of the two optical paths in the parent MZI and a Quadrature MZI is arranged in the other optical path. The optical path difference of the In-Phase MZI is set to ±λ / 2×(2m I +1) so that the light with wavelength λ is extinguished at the moment when the modulation signal becomes 0. The optical path difference of the Quadrature MZI is set to ±λ / 2×(2m Q +1) so that the light with wavelength λ is extinguished at the moment when the modulation signal becomes 0. m I and m Q are different integers of 0 or more, respectively.

[0137] The optical circuit may be an IQ optical modulator having an in-line MZI in which an In-Phase MZI is arranged in one of the two optical paths in the parent MZI and a Quadrature MZI is arranged in the other optical path. The optical path difference of the In-Phase MZI and the optical path difference of the Quadrature MZI are set to an odd multiple of ±λ / 2 so that light of wavelength λ is extinguished at the moment when the modulation signal becomes 0. In this case, the controller controls to correct the optical path difference of the In-Phase MZI or the optical path difference of the Quadrature MZI so that the optical intensity P(λ) of the propagated input light of wavelength λ output from the optical circuit becomes small or large.

[0138] As described above, the embodiments of the present invention have been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and designs and the like within the scope not departing from the gist of the present invention are also included.

Description of Reference Numerals

[0139] 1 Optical input port 2 Optical branching circuit 3 First optical waveguide 4 Second optical waveguide 5 Bias application electrode 6 Optical multiplexing circuit 7 Output port 8 Wavelength division multiplexing coupler 10 Optical tap circuit 11 Optical power monitor 12 Optical bandpass filter 13 Wavelength separation coupler 20 I-side optical branching circuit 21 I-side optical multiplexing circuit 22 I-side modulation electrode 23 I-side bias application electrode 24 Q-side optical branching circuit 25 Q-side optical multiplexing circuit 26 Q-side modulation electrode 27 Q-side bias application electrode 32 Wavelength tunable light source 33 Optical coupler 34 Synchronous Detection / Averaging Processing Circuit 35-1 First Reference Light Source 35-2 Second Reference Light Source 41 Controller 42 Controller 43 Controller 44 Controller 45 Controller 50 MZI for In-Phase 51 MZI for Quadrature 52 Optical Power Monitor 53-1 First Reference Optical Power Monitor 53-2 Second Reference Optical Power Monitor 61 Wavelength Division Coupler 62 Wavelength Multiplexing Coupler 91 MZI 92 IQ Optical Modulator 101 Optical Output Circuit 102 Optical Output Circuit 103 Optical Output Circuit 104 Optical Output Circuit 105 Optical Output Circuit

Claims

1. A control circuit for controlling at least one optical path difference among the optical path differences of two optical paths respectively included in one or more Mach-Zehnder interferometers included in an optical circuit, When N is a natural number, N types of wavelengths λ different from the wavelength λ of the input light input to the optical circuit １ ~λ Ｎ A reference light generation unit that generates each of the reference lights in parallel or in a time-division manner, A reference light input unit that multiplexes each of the N types of wavelengths λ1 to λN of the reference lights, or multiplexes the reference light and the input light and inputs the multiplexed light to the optical circuit, The light intensity of the propagated input light that is the input light after propagating through one or more of the Mach-Zehnder interferometers included in the optical circuit, the light intensity of the propagated reference light that is the reference light after propagating through one or more of the Mach-Zehnder interferometers through which the propagated input light has propagated, or the light intensity of the light obtained by multiplexing the propagated input light and the propagated reference light, and an optical power monitor unit that monitors the light intensity, The light intensity P(λ) of the propagated input light of wavelength λ obtained using the monitoring result by the optical power monitor unit and N types of wavelengths λ １ ~λ Ｎ The light intensity P(λ １ )~P(λ Ｎ ) of each of the propagated reference lights, and a controller that controls to correct the optical path difference based on the above, A control circuit comprising the above.

2. A control circuit for controlling at least one optical path difference among the optical path differences of two optical paths respectively included in one or more Mach-Zehnder interferometers included in an optical circuit, When N is a natural number, N types of wavelengths λ different from the wavelength λ of the input light input to the optical circuit １ ~λ Ｎ A reference light generation unit that generates each of the reference lights in parallel or in a time-division manner, A reference light input unit that inputs the reference light to the optical circuit, The optical power monitor unit that monitors the optical intensity of the propagated input light, which is the input light after propagating through the single or multiple Mach-Zehnder interferometers included in the optical circuit, the optical intensity of the propagated reference light, which is the reference light after propagating through the single or multiple Mach-Zehnder interferometers through which the propagated input light has propagated, or the optical intensity of the light obtained by multiplexing the propagated input light and the propagated reference light, N types of wavelengths λ obtained using the monitoring result by the optical power monitor unit １ ~λ Ｎ Based on the optical intensity P(λ １ )~P(λ Ｎ ) of each of the propagated reference lights, a controller that controls to correct the optical path difference, A control circuit comprising:

3. The reference light input unit multiplexes the N types of reference lights with the input light and inputs the multiplexed light into the optical circuit, The control circuit further includes a wavelength separation unit that separates the propagated input light and the N types of propagated reference lights output in multiplex from the optical circuit, The optical power monitor unit detects the optical intensity of each of the propagated input light and the N types of propagated reference lights separated by the wavelength separation unit, The control circuit according to claim 1 or claim 2.

4. The controller controls the extinction of the reference light and the generation of the reference light by time division in the reference light generation unit, The reference light input unit multiplexes the reference light with the input light and inputs the multiplexed light into the optical circuit, When n is a natural number from 1 to N, the optical power monitor unit outputs the optical intensity P(λ) obtained by monitoring the propagated input light of wavelength λ and the optical intensity P(λ,λ ｎ ) obtained by monitoring the light obtained by multiplexing the propagated input light of wavelength λ and the propagated reference light of wavelength λ ｎ ) to the controller, The controller, based on the optical intensity P(λ) and the optical intensity P(λ,λ ｎ ), for wavelengths λ １ ~λ ＮThe optical intensity P(λ １ ) to P(λ Ｎ ) of each of the reference lights after propagation is calculated. The control circuit according to claim 1.

5. The controller controls the generation of the reference light by time division in the reference light generation unit. The reference light input unit multiplexes the reference light with the input light and inputs the multiplexed light into the optical circuit. When n is a natural number from 1 to N, the optical power monitor unit monitors the light obtained by multiplexing the input light after propagation with wavelength λ and the reference light after propagation with wavelength λ ｎ and outputs the optical intensity P(λ, λ ｎ ) thus obtained to the controller. The controller calculates the optical intensity P(λ, λ ｎ ) of each of the reference lights after propagation with wavelengths from λ １ to λ Ｎ based on the optical intensity P(λ, λ １ ) to P(λ Ｎ ). The control circuit according to claim 2.

6. The controller controls the extinction of the reference light and the generation of the optical pulse train of the reference light by a pulse train in the reference light generation unit. The reference light input unit multiplexes the optical pulse train of the reference light generated by time division with the input light and inputs the multiplexed light into the optical circuit. When n is a natural number from 1 to N, the control circuit averages the results of monitoring, by the optical power monitor unit, the light output from the optical circuit while the reference light generation unit is in the extinction state to obtain the optical intensity P(λ) of the input light after propagation with wavelength λ, and synchronously detects, by a reference clock used for the generation of the optical pulse train, the results of monitoring, by the optical power monitor unit, the light output from the optical circuit while the reference light generation unit is generating the optical pulse train of the reference light with wavelength λ ｎ to obtain the optical intensity P(λ ｎ ) of the reference light after propagation with wavelength λ ｎ and further includes a synchronous detection unit. The control circuit according to claim 1.

7. The reference light generated by the reference light generation unit is an optical pulse train, The controller controls extinction of the reference light and generation of the optical pulse train in the reference light generation unit, The period and phase of the optical pulse train of the reference light are determined by a reference clock generated by the controller, The reference light input unit multiplexes the optical pulse train of the reference light generated in a time-division manner with the input light and inputs the multiplexed light into the optical circuit, When n is a natural number from 1 to N, the control circuit averages the results of monitoring, by the optical power monitor unit, of the light output from the optical circuit while the reference light generation unit is in extinction, to obtain the optical intensity P(λ) of the propagated input light with wavelength λ, and synchronously detects, by the reference clock used for generation of the optical pulse train, the results of monitoring, by the optical power monitor unit, of the light output from the optical circuit while the reference light generation unit is generating the optical pulse train of the reference light with wavelength λ, to obtain the optical intensity P(λ ｎ ) of the propagated reference light, and further includes a synchronous detection unit. ｎ of the propagated reference light, ｎ ) The control circuit according to claim 1.

8. The reference light generated by the reference light generation unit is an optical pulse train, The controller controls generation of the optical pulse train of the reference light in the reference light generation unit, The period and phase of the optical pulse train of the reference light are determined by a reference clock generated by the controller, The reference light input unit multiplexes the optical pulse train of the reference light generated in a time-division manner with the input light and inputs the multiplexed light into the optical circuit, When n is a natural number from 1 to N, the control circuit synchronously detects, by the reference clock used for generation of the optical pulse train, the results of monitoring, by the optical power monitor unit, of the light output from the optical circuit while the reference light generation unit is generating the optical pulse train of the reference light with wavelength λ, to obtain the optical intensity P(λ ｎ of the propagated reference light, ｎThe optical intensity P(λ of the reference light after the propagation ｎ ), and further includes a synchronous detection unit for obtaining the control circuit according to claim 2.

9. The reference light propagates through the single or a plurality of Mach-Zehnder interferometers through which the input light has propagated in a direction opposite to that of the input light. The control circuit according to claim 1 or claim 2.

10. The interval between each of the wavelengths of the N types of reference lights and the wavelength of the input light does not coincide with any of the free spectral ranges of the single or a plurality of Mach-Zehnder interferometers of the optical circuit. The control circuit according to claim 1 or claim 2.

11. In any one of the single or a plurality of Mach-Zehnder interferometers of the optical circuit, when the optical path difference is set so that the interference intensity of the input light with wavelength λ is maximized, if k is an integer other than 0, the optical path difference is k times the wavelength λ. The control circuit according to claim 1 or claim 2.

12. The optical circuit is an optical modulator, A modulation signal is applied to at least one of the single or a plurality of Mach-Zehnder interferometers of the optical circuit. The control circuit according to claim 1 or claim 2.

13. The optical circuit is an IQ optical modulator having an nested Mach-Zehnder interferometer in which a Mach-Zehnder interferometer for in-phase is arranged in one of the two optical paths in the parent Mach-Zehnder interferometer, and a Mach-Zehnder interferometer for quadrature phase is arranged in the other optical path. The Mach-Zehnder interferometer for in-phase has an optical path difference set to ±λ / 2×(2m Ｉ + 1) so that light with wavelength λ is extinguished when the modulation signal is 0. The Mach-Zehnder interferometer for quadrature phase has an optical path difference set to ±λ / 2×(2m Ｑ + 1) so that light with wavelength λ is extinguished when the modulation signal is 0. The m Ｉ and m Ｑ are different integers equal to or greater than 0, The control circuit according to claim 12.

14. The optical circuit is an IQ optical modulator having a nested Mach-Zehnder interferometer in which a Mach-Zehnder interferometer for in-phase is arranged in one of the two optical paths in the parent Mach-Zehnder interferometer and a Mach-Zehnder interferometer for quadrature phase is arranged in the other optical path, In the in-phase Mach-Zehnder interferometer and the quadrature-phase Mach-Zehnder interferometer, the optical path difference of the in-phase Mach-Zehnder interferometer and the optical path difference of the quadrature-phase Mach-Zehnder interferometer are set to an odd multiple of ±λ / 2 so that light of wavelength λ is extinguished when the modulation signal is 0, The controller controls to correct the optical path difference of the in-phase Mach-Zehnder interferometer or the optical path difference of the quadrature-phase Mach-Zehnder interferometer so that the light intensity P(λ) of the propagated input light of wavelength λ output from the optical circuit becomes small or large, The control circuit according to claim 12.

15. An optical circuit control method for controlling at least one of the optical path differences of two optical paths respectively included in a single or a plurality of Mach-Zehnder interferometers included in an optical circuit, When N is a natural number, N types of wavelengths λ １ ~λ Ｎ A reference light generation step of generating each reference light in parallel or in a time-division manner, A reference light input step of inputting the reference light into the optical circuit, A monitoring step of monitoring the light intensity of the propagated input light, which is the input light after propagating through the single or a plurality of Mach-Zehnder interferometers included in the optical circuit, the light intensity of the propagated reference light, which is the reference light after propagating through the single or a plurality of Mach-Zehnder interferometers through which the propagated input light has propagated, or the light intensity of the light obtained by multiplexing the propagated input light and the propagated reference light, The optical intensity P(λ) of the input light after propagation at wavelength λ obtained using the monitoring result in the monitoring step, and the N types of wavelengths λ １ ~λ Ｎ The optical intensity P(λ １ )~P(λ Ｎ ) of each of the reference lights after propagation, or, without using the optical intensity P(λ), the optical intensity P(λ １ )~P(λ Ｎ ) to control so as to correct the optical path difference based on; and a control step An optical circuit control method having the above.

Images