JP2011039405A - Optical signal transducing apparatus and method of controlling the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an optical signal transducing apparatus which is stably controlled. <P>SOLUTION: The optical signal transducing apparatus includes: a plurality of input ports; a plurality of output ports; a plurality of mirrors M1 to Mn which reflect an optical signal input from the input ports and output it to any one of the output ports; a control means 110 which controls the reflection angle of the mirrors to select the path of the optical signals. The control means includes: a generation means which generates a low frequency signal having a predetermined frequency; a superimposing means which superimposes the low frequency signal upon the control signal which controls the reflection angle of the mirrors; a photoelectric conversion means which converts part of the optical signal output from the output ports to an electric signal; a logarithmic conversion means which logarithmically converts the obtained electric signal; an extraction means which performs the multiplication of the logarithmically transformed electric signal and the low frequency signal similar to the superimposed low frequency signal, transforms a signal component having a predetermined frequency into a DC component to extract it; and a generation means which generates a control signal for setting the mirrors at a predetermined reflection angle on the basis of the extracted signal component. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an optical signal switching device and a method for controlling the optical signal switching device.

In an optical network using a wavelength division multiplexing (WDM) system in recent years, an optical signal exchange is performed in which each wavelength path of a multiplexed signal is switched as an optical signal in order to form a more flexible network. The importance of equipment is increasing.
As a means for realizing such an optical signal exchange device, for example, a device using a movable mirror formed by MEMS (Micro Electric Mechanical System) technology has been proposed, which is promising from the viewpoint of optical characteristics and expandability. It is considered as a method.

  FIG. 14 shows a configuration example of a conventional optical signal switching apparatus 1 using a MEMS mirror. The WDM signals input from the optical fibers 2-1 to 2-n constituting the input port are output from the fiber collimators 3-1 to 3-n, the paths are separated for each wavelength by the diffraction grating 4, and each wavelength is supported. The light is condensed on the MEMS mirrors M <b> 1 to Mn through the lens 5. The signals switched by the MEMS mirrors M1 to Mn and switched in their paths are incident on the fiber collimators 6-1 to 6-n corresponding to the desired output ports, and the optical fibers 8-1 to 8- output from n.

  FIG. 15 shows the relationship between the angle (θ) of the MEMS mirrors M1 to Mn and the optical coupling efficiency (h) of the fiber collimators 6-1 to 6-n. The optical coupling efficiency between the MEMS mirrors M1 to Mn and the fiber collimators 6-1 to 6-n has the following Gaussian having a peak at an angle at which the light beam is incident on the center of the fiber collimators 6-1 to 6-n. It can be expressed as a function.

Where θ is the angle of the MEMS mirror, θ ₀ is the optimum angle at which the optical coupling efficiency is maximized, and σ is the coupling radius with the fiber collimators 6-1 to 6-n on the output side, depending on the wavelength, beam diameter, etc. The value to be determined. As shown in FIG. 15, in the case of an apparatus using a MEMS mirror, there is a certain optimum mirror angle for optical coupling with a fiber collimator. However, even if the optical coupling is optimally adjusted in the initial adjustment, the mirror angle changes and the optical coupling characteristics to the fiber collimator deteriorate due to factors such as ambient temperature fluctuation, vibration, and aging degradation. There is. Therefore, for example, as shown in FIG. 14, the optical power of the optical signal output from the fiber collimator is monitored by the monitor circuit 10, and the voltage applied to the mirror electrode is controlled based on the monitor information. It is necessary to adjust the angle of the mirror so that the optical coupling to the output side fiber collimators 6-1 to 6-n is optimized.

  As a method of compensating for the mirror angle, a low frequency signal (Dither) is added to the voltage applied to the mirror, and a signal component having the same frequency as the low frequency signal included in the optical output according to the low frequency signal (hereinafter, A method is proposed in which a mirror angle is compensated by detecting a deviation from the optimum angle of optical coupling in the fiber collimators 6-1 to 6-n based on the result. (See Patent Documents 1 and 2).

  FIG. 16 shows a configuration part of a control circuit related to MEMS mirror angle compensation in the prior art. In this example, the control circuit 11 includes a low frequency component extraction unit 20, a control signal generation unit 21, a low frequency signal superimposition unit 22, and a low frequency signal generation unit 23. The low frequency signal generator 23 generates a low frequency signal having a predetermined frequency. The low-frequency signal generated here is added to the control signal by the low-frequency signal superimposing unit 22 and output, and the voltage including the low-frequency signal is applied to the MEMS mirrors M1 to Mn by the driver circuit 12 shown in FIG. Apply. Accordingly, the MEMS mirrors M1 to Mn slightly vibrate according to the low frequency signal, and the reflection angle changes.

  For this reason, the optical signal reflected from the MEMS mirrors M1 to Mn includes the frequency component (low frequency component) of the added low frequency signal. The optical signal including the low frequency component is detected by the monitor circuit 10 shown in FIG. The low frequency component extraction unit 20 multiplies the signal of the optical monitor input by the low frequency signal generated by the low frequency signal generation unit 23, and extracts only the DC component by a low pass filter (hereinafter referred to as LPF). The low frequency component contained in the optical signal is extracted. Furthermore, the control signal generation unit 21 performs feedback control using the extracted low frequency component so that the low frequency component becomes zero, such as proportional / integral / derivative control (PID control), and a new control signal. Is generated.

  FIG. 17 shows the state of low-frequency components included in the optical output when the optical output is off the peak (a) and when the optical output is near the peak (b). Even when a low-frequency signal having the same amplitude is applied, the low-frequency component included in the light output becomes smaller as the light output is closer to the peak. That is, as described above, the mirror angle is compensated so that the optical coupling to the fiber collimators 6-1 to 6-n is optimized by bringing the low frequency component included in the light output close to zero.

JP 2003-029175 A US Pat. No. 6,753,960

  In the prior art, the optical output power is received by a photodiode, the output is amplified by an amplifier, and a value proportional to the optical output power is captured as a monitor signal (Patent Document 1, Paragraph 0039 and Patent Document 2). Formula (7)). In other words, the detected monitor signal has a value proportional to the Gaussian function expressed by Equation (1). On the other hand, when the above-described low frequency component detection method is used, a value proportional to the first derivative with respect to the angle of the optical monitor signal is the extracted low frequency component (Equations (5) and (7) of Patent Document 2). ). That is, in the prior art, a value proportional to the first derivative with respect to the mirror angle of the Gaussian function expressed by Equation (1) is extracted as the low frequency component included in the light output.

  FIG. 18 shows the angle dependency of the extracted low frequency component in the prior art. The low frequency component has a value proportional to the first derivative of the Gaussian function, and has a maximum value or a minimum value at a position shifted from the peak position by the coupling radius. In this case, as shown in FIG. 18, when optical coupling is performed at a position closer to the peak than the coupling radius, the low-frequency component, that is, the first-order differential value of the Gaussian function is controlled to be small. As shown by the solid line arrow, it can be converged to a peak, and operates normally as angle compensation of the MEMS mirror.

  However, for example, when the MEMS mirror moves greatly due to vibration or the like and the deviation becomes larger than the coupling radius, the low frequency component is proportional to the first derivative of Gaussian. Then, it is controlled in a direction away from the peak, and the optical coupling with the fiber collimator diverges as shown by the broken line arrow in FIG.

As described above, in the related art, it is difficult to perform normal control because it is not possible to recover when the angle greatly deviates from the optimum angle due to large vibration or the like. Further, since the monitored optical output power is a value obtained by multiplying the optical input power and the optical coupling efficiency of the fiber collimator, if the optical input power is P ₀ and the monitored optical output power is P (θ), It is expressed by the following formula.

That is, the monitored optical output power depends not only on the optical coupling efficiency to the fiber collimator but also on the optical input power. Further, since the extracted low-frequency component is a value proportional to the first derivative with respect to the mirror angle in Expression (2), the low-frequency component is also a value depending on the optical input power. That is, when the first-order differentiation of Equation (2) with respect to θ, a term of P ₀ · θ is generated, so that the low frequency component has dependency on P ₀ .

FIG. 19 shows the extracted low frequency components when the optical input power is large (curve indicated by a solid line) and small (curve indicated by a broken line). As shown in FIG. 19, when the optical input power is large and the deviation from the optimum angle is small (right thin line (θ ₁ )), and when the optical input power is small and the deviation from the optimum angle is large (left fine line). (Θ ₂ )), a low frequency component having the same value (dashed thin line (D ₀ )) is extracted. In such a case, since the respective states cannot be distinguished, the optical coupling efficiency (that is, the deviation from the peak) with the fiber collimator cannot be accurately detected in the conventional method.

  When performing mirror compensation, if the deviation from the optimum angle is large, the compensation amount of the voltage applied to the mirror can be increased, and if the compensation amount can be reduced as the optimum angle is approached, stable and high speed can be achieved. Control can be performed. However, since the prior art cannot accurately detect a deviation from the correct optimum angle, convergence to the optimum angle is delayed when the optical input power is small, or conversely, the optical output oscillates and is not effective when the optical input power is large. It becomes difficult to perform stable and high-speed control such as becoming stable.

  Therefore, the problem to be solved by the present invention is to provide an optical signal switching apparatus capable of stable control and a method for controlling the optical signal switching apparatus.

In order to solve the above problems, an optical signal switching device according to the present invention includes a plurality of input ports to which an optical signal is input, a plurality of output ports to which an optical signal is output, and light input from the predetermined input port. A plurality of mirrors that reflect a signal and guide it to the desired output port; and a control unit that controls a reflection angle of the mirror and selects a propagation path of the optical signal between the input port and the output port. The control means includes a generating means for generating a low-frequency signal having a predetermined frequency, a superimposing means for superposing the low-frequency signal on a control signal for controlling the reflection angle of the mirror, and an output port. Photoelectric conversion means for converting a part of the output optical signal into an electric signal, logarithmic conversion means for logarithmically converting the electric signal obtained by the photoelectric conversion means, and electric power output from the logarithmic conversion means Extraction means for extracting the signal component of the predetermined frequency included in the signal, and generation means for generating a control signal for setting the mirror to a desired reflection angle according to the extracted signal component. It is characterized by.
According to such a configuration, an optical signal switching device capable of stable control can be provided.

In another aspect of the invention, in addition to the above-described invention, the optical signal is wavelength-multiplexed light, and the wavelength-multiplexed light input from the input port is separated into optical signals of respective wavelengths and then incident on the mirrors. And the control means controls the reflection angle of each mirror so that an optical signal of each wavelength is output from a desired output port.
According to such a configuration, it is possible to exchange optical signals for wavelength multiplexed light.

Another invention is characterized in that, in addition to the above-mentioned invention, it has a high-pass means for attenuating a direct current component from an electric signal output from the logarithmic conversion means and passing a signal component of other frequency. .
According to such a configuration, the influence of noise can be reduced and the mirror can be reliably controlled.

In addition to the above-mentioned invention, another invention has a band-pass means for passing the signal component of the predetermined frequency and attenuating the signal component of the other frequency from the electrical signal output from the logarithmic conversion means. And
According to such a configuration, the influence of noise can be reduced and the mirror can be reliably controlled.

Another invention includes a calculation unit that calculates an attenuation amount of the optical signal in accordance with a signal component of the predetermined frequency included in the electrical signal output from the logarithmic conversion unit, and the control unit includes the calculation The angle of the mirror is controlled so that the attenuation calculated by the means becomes equal to the target attenuation.
According to such a configuration, it is possible to accurately control the attenuation amount to be a desired value.

In another aspect of the invention, in addition to the above-described invention, the mirror can independently control a reflection angle about two axes orthogonal to each other, and the control means can control either one of the two axes. The low frequency signal is superimposed on a control signal for controlling the reflection angle centered, and based on the result, control is performed so that the reflection angle centered on the target axis becomes a desired angle. To do.
According to such a configuration, the reflection angle can be adjusted for each of the two directions of the mirror.

In another aspect of the invention, in addition to the above-described invention, the mirror can independently adjust the reflection angle about two axes orthogonal to each other, and the control means is a reflection centered on the two axes. Superimposing a low-frequency signal with a phase difference of π / 2 on each control signal for controlling the angle, and controlling the reflection angle of each axis to a desired angle based on these low-frequency signals It is characterized by.
According to such a configuration, it is possible to simultaneously adjust the reflection angle in each of the two directions of the mirror.

In another invention, in addition to the above invention, the control means includes the generation means, the superimposition means, the extraction means, and the generation means in a number corresponding to the mirror to be controlled, and the generation The means generates a low-frequency signal having a predetermined frequency different for each mirror, the superimposing means superimposes the low-frequency signal on a control signal to each mirror, and the logarithmic conversion means is output from the photoelectric conversion means An electrical signal is logarithmically converted, each of the extraction means extracts a signal component of the predetermined frequency of each mirror included in the electrical signal output from the logarithmic conversion means, and each of the generation means Each mirror is controlled to a desired reflection angle according to the component.
According to such a configuration, the angles of the plurality of mirrors can be adjusted simultaneously.

The optical signal switching device control method of the present invention reflects a plurality of input ports to which an optical signal is input, a plurality of output ports to which an optical signal is output, and an optical signal input from a predetermined input port. A plurality of mirrors that lead to the desired output port, and a control unit that controls a reflection angle of the mirror and selects a propagation path of the optical signal between the input port and the output port. In the control method, the control means generates a low frequency signal having a predetermined frequency, a superimposing step of superimposing the low frequency signal on a control signal for controlling a reflection angle of the mirror, A photoelectric conversion step for converting a part of the optical signal output from the output port into an electrical signal, and a logarithmic conversion step for logarithmically converting the electrical signal obtained in the photoelectric conversion step. An extraction step for extracting the signal component of the predetermined frequency included in the electrical signal output from the logarithmic conversion step, and setting the mirror to a desired reflection angle according to the extracted signal component And a generation step of generating a control signal.
According to such a configuration, it is possible to provide a method for controlling an optical signal switching apparatus capable of stable control.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to provide the control method of the optical signal switching apparatus which can be controlled stably, and an optical signal switching apparatus.

It is a figure which shows the structural example of the optical signal switching apparatus which concerns on embodiment of this invention. FIG. 2 is a block diagram illustrating a configuration example according to a mirror angle compensation control circuit of the control circuit illustrated in FIG. 1. It is a figure which shows the angle dependence of the low frequency component detected in embodiment shown in FIG. It is a figure which shows the detailed structural example of the MEMS mirror used in 2nd Embodiment which concerns on the mirror angle compensation control circuit of a control circuit. It is a block diagram which shows 3rd Embodiment which concerns on the mirror angle compensation control circuit of a control circuit. It is an example of the power spectrum of the signal output from the low frequency component extraction part in embodiment shown in FIG. It is an example of the power spectrum of the signal output in embodiment shown in FIG. It is a block diagram which shows 4th Embodiment which concerns on the mirror angle compensation control circuit of a control circuit. It is an example of the power spectrum of the signal output in embodiment shown in FIG. It is a block diagram which shows 5th Embodiment which concerns on the mirror angle compensation control circuit of a control circuit. It is a block diagram which shows 6th Embodiment which concerns on the mirror angle compensation control circuit of a control circuit. It is a block diagram which shows 7th Embodiment which concerns on the mirror angle compensation control circuit of a control circuit. It is a figure which shows the structural example of the conventional loss constant control circuit. It is a figure which shows the structural example of the conventional optical signal switching apparatus. It is a figure which shows the relationship between the mirror angle and the optical coupling efficiency of a fiber collimator. FIG. 15 is a block diagram illustrating a configuration example according to a mirror angle compensation control circuit of the control circuit illustrated in FIG. 14. It is a figure which shows the mode of the low frequency component contained in an optical output. It is a figure which shows the angle dependence of the low frequency component detected in the optical signal switching apparatus shown in FIG. It is a figure which shows the optical input power dependence of the low frequency component detected in the optical signal switching apparatus shown in FIG.

  Next, an embodiment of the present invention will be described.

(A) First embodiment

  FIG. 1 is a diagram showing a configuration example of an optical signal switching apparatus according to the present invention. As shown in this figure, the optical signal switching device 1 includes optical fibers 2-1 to 2-n, fiber collimators 3-1 to 3-n, a diffraction grating 4, a condenser lens 5, MEMS mirrors M1 to Mn, and a fiber collimator. 6-1 to 6-n, optical couplers 7-1 to 7-n, optical fibers 8-1 to 8-n, a monitor circuit 10, a control circuit 110, and a driver circuit 12. The optical fibers 2-1 to 2-n and the fiber collimators 3-1 to 3-n constitute input ports, and the fiber collimators 6-1 to 6-n and the optical fibers 8-1 to 8-n constitute output ports. To do.

  Here, the optical fibers 2-1 to 2-n are constituted by n optical fibers, and fiber collimators 3-1 to 3-n are connected to one end thereof. A WDM signal is input to the other ends of the optical fibers 2-1 to 2-n. The fiber collimators 3-1 to 3-n are modules in which an optical fiber and a lens are integrated, and have a function of emitting a parallel beam.

  The diffraction grating 4 reflects and separates the light beam (WDM light) emitted from the fiber collimators 3-1 to 3-n at an angle according to the wavelength of the included light beam, and then collects the condensing lens 5. Is incident on. The condensing lens 5 condenses the light of each wavelength separated by the diffraction grating 4 and enters the MEMS mirrors M1 to Mn corresponding to the respective wavelengths.

  The MEMS mirrors M1 to Mn each independently control the angles in the X-axis direction and the Y-axis direction by a drive signal output from the driver circuit 12, and reflect the incident light beam in a predetermined direction. At this time, the reflection surface of each MEMS mirror is controlled to a predetermined angle corresponding to the position of any one of the fiber collimators 6-1 to 6-n set as the output destination of the incident light beam. The light beams of the respective wavelengths reflected by the MEMS mirrors enter the target fiber collimators 6-1 to 6-n. More specifically, the light beams reflected by the MEMS mirrors M1 to Mn enter the desired fiber collimators 6-1 to 6-n via the condenser lens 5 and the diffraction grating 4. In FIG. 1, the direction in which the light beam of each wavelength is angularly dispersed by the diffraction grating 4 is the X direction, the direction in which the input / output ports are arranged is the Y direction, and the direction perpendicular to the XY plane is the Z direction. It is said.

  The optical couplers 7-1 to 7-n branch a part of the light beam incident on the fiber collimators 6-1 to 6-n (for example, about 1/100 of the incident light) and supply the light to the monitor circuit 10. Output. The optical fibers 8-1 to 8-n are composed of n optical fibers that transmit the light beams output from the optical couplers 7-1 to 7-n.

  The monitor circuit 10 photoelectrically converts the light beams output from the optical couplers 7-1 to 7-n, and outputs a signal having a voltage (or current) corresponding to the intensity. The control circuit 110 generates a control signal for controlling the MEMS mirrors M <b> 1 to Mn based on a signal (optical monitor signal) output from the monitor circuit 10, and outputs the control signal to the driver circuit 12. The driver circuit 12 amplifies the power of the control signal supplied from the control circuit 110 and applies it to the MEMS mirrors M1 to Mn to control the reflection angle of each MEMS mirror.

  In the optical signal switching apparatus 1 according to the embodiment shown in FIG. 1, light having an arbitrary wavelength is controlled by controlling the reflection angles of the MEMS mirrors M <b> 1 to Mn with respect to light having a plurality of wavelengths included in the input WDM light. The beam can be selected and directed to the desired output port.

  FIG. 2 is a block diagram showing a configuration example of the control circuit 110 shown in FIG. In this figure, only the portion related to the “mirror angle control correction unit” for correcting the single MEMS mirror to have an optimum angle is shown in the control circuit 110. As shown in this figure, the control circuit 110 includes a logarithmic conversion unit 120, a low frequency component extraction unit 121, a control signal generation unit 122, a low frequency signal superimposition unit 123, and a low frequency signal generation unit 124 as main components. It is said.

  Here, the logarithmic conversion unit 120 receives a signal (optical monitor signal) output from the monitor circuit 10, calculates the logarithm of the signal value, and outputs it. The low frequency component extraction unit 121 includes a multiplication circuit 121a and an LPF (Low Pass Filter) 121b, and extracts and outputs a low frequency component from the signal output from the logarithmic conversion unit 120. More specifically, the multiplier circuit 121a adds a low-frequency signal (an angular frequency of ω = 2πf (where f is a frequency)) supplied from the low-frequency signal generator 124 to the signal output from the logarithmic converter 120. ), The signal having the angular frequency ω included in the signal output from the logarithmic converter 120 is converted into a DC component and output. The LPF 121b passes the DC component from the signal included in the signal output from the multiplication circuit 121a and attenuates and outputs the other components. The DC component output from the LPF 121b is a signal proportional to the amount of deviation from the target angle of the MEMS mirror (details will be described later). Here, ω is set to a frequency sufficiently lower than the resonance frequency of the MEMS mirror so that the MEMS mirror does not resonate due to the superimposed low-frequency signal. For example, when the natural resonance frequency of the MEMS mirror is several kHz, the frequency of the low frequency signal can be set to a frequency of about 1/10, for example, about 100 Hz.

  The control signal generation unit 122 generates and outputs a control signal for controlling each MEMS mirror according to the low frequency component that is converted into the direct current component by the low frequency component extraction unit 121 and extracted. The control signal generation unit 122 generates a control signal for adjusting the angle so that the reflected light is incident on a desired output port by the MEMS mirror to be controlled, and outputs the control signal from the low frequency component extraction unit 121. The control signal is corrected and output in order to correct the deviation from the optimum angle of the MEMS mirror in accordance with the signal to be output. The low frequency signal superimposing unit 123 superimposes and outputs the low frequency signal output from the low frequency signal generating unit 124 on the control signal generated by the control signal generating unit 122. The control signal on which the low-frequency signal output from the low-frequency signal superimposing unit 123 is superimposed is supplied to the driver circuit 12 and is supplied to the driving unit of each MEMS mirror, and each angle is controlled.

  Next, the operation of the first embodiment will be described.

Hereinafter, to simplify the description, paying attention to the MEMS mirror M1, among the WDM light input from the optical fiber 2-1, the light beam of wavelength lambda ₁ is reflected is incident on the MEMS mirror M1, the optical fiber An example of outputting from 8-n will be described. Further, the control axis of the MEMS mirror M1 will be described by taking only one direction (for example, only the X-axis direction) as an example for simplification of description.

When WDM light is input to the optical fiber 2-1, it is converted into parallel rays by the fiber collimator 3-1 and then incident on the diffraction grating 4. The diffraction grating 4 reflects the light beam at an angle corresponding to each wavelength included in the WDM light. As a result, the light beam having the wavelength λ ₁ is incident on the MEMS mirror M 1 via the condenser lens 5.

The angle of the MEMS mirror M1 is set so that the reflected light is incident on the fiber collimator 6-n via the condenser lens 5 and the diffraction grating 4. Therefore, of the WDM light input to the optical fiber 2-1, the light beam having a wavelength of λ ₁ is reflected after being incident on the MEMS mirror M 1 by the diffraction grating 4 and the condenser lens 5, and the condenser lens 5 and The light enters the fiber collimator 6-n via the diffraction grating 4 and is output from the output port 8-n.

  By the way, the angle of the MEMS mirror M1 may deviate from the optimum value due to, for example, secular change or environmental temperature. In such a case, a part of the light beam reflected by the MEMS mirror M1 is not incident on the fiber collimator 6-n, so that the level of the output optical signal is lowered.

  Therefore, in this embodiment, control is performed so that the MEMS mirror M1 has an optimum angle regardless of aging and environmental temperature. More specifically, in the control circuit 110, a signal whose angular frequency is ω output from the low-frequency signal generator 124 (for example, cos (ωt)) with respect to the control signal output from the control signal generator 122. Are superimposed and output. Here, t indicates time.

  When a control signal on which such a low frequency signal is superimposed is supplied to the drive unit of the MEMS mirror M1, for example, in the X-axis direction, the MEMS mirror M1 is centered on the current position (set position) and is low. Microvibration according to the frequency signal. As a result, since the reflection angle by the MEMS mirror M1 changes according to minute vibrations, the incident position of the light beam on the fiber collimator 6-n changes (vibrates) according to the low frequency signal.

  A part of the light beam incident on the fiber collimator 6-n is branched by the optical coupler 7-n and guided to the monitor circuit 10. In the monitor circuit 10, the light from the optical coupler 7-n is photoelectrically converted into a signal having a voltage value (or current value) corresponding to the intensity of the light beam, and is output to the control circuit 110.

The signal input to the control circuit 110 is logarithmically converted by the logarithmic converter 120 and output. Now, when the low frequency signal is not superimposed, the relationship between the angle θ, the optimum angle θ ₀ , the coupling radius σ, and the optical input power P ₀ of the MEMS mirror M1 is expressed by the above-described equation (2). . Therefore, the signal output from the logarithmic conversion unit 120 is expressed by the following equation (3). A is a proportionality coefficient determined by the monitor circuit 10 or the like.

Here, the angle θ of the MEMS mirror M1 to which the low-frequency signal is applied is expressed by the following formula (4). Θ _i is the current angle of the MEMS mirror M1, and Δθ and ω are the amplitude and angular frequency of the angle of the MEMS mirror M1 due to the low-frequency signal.

  When Expression (4) is substituted into Expression (3), the following Expression (5) indicating the output signal from the logarithmic conversion unit 120 when the low frequency signal is superimposed is obtained.

  The signal output from the logarithmic conversion unit 120 is supplied to the low frequency component extraction unit 121. In the low frequency component extraction unit 121, the low frequency component is converted into a corresponding DC component and output. Specifically, the multiplication circuit 121 a of the low frequency component extraction unit 121 multiplies the signal output from the logarithmic conversion unit 120 by the low frequency signal cos (ωt) output from the low frequency signal generation unit 124. . When the above equation (5) is multiplied by cos (ωt), the following equation (6) is obtained.

  The signal output from the multiplication circuit 121a is supplied to the LPF 121b, a DC component is extracted, and the other components are attenuated. More specifically, in the equation (6), a signal related to a term including cos (ωt), cos (2ωt), and cos (3ωt) is attenuated, and only a DC component shown in the following equation (7) is obtained. Is output.

Equation (7) is a value that depends only on the deviation (θ _i −θ ₀ ) from the optimum angle, and does not depend on the optical input power P ₀ as in the conventional example. This makes it possible to accurately detect a deviation from the optimum angle. Such a signal is supplied from the LPF 121b to the control signal generator 122. The control signal generation unit 122 finely adjusts the angle of the MEMS mirror M1 based on the signal supplied from the LPF 121b so that the value of the signal becomes small (so that θ _i = θ ₀ ). This makes it possible to set the MEMS mirror M1 to the optimum angle theta _0.

That is, even if the optical input power P ₀ is different, if the deviation from the optimum angle is the same, the same value is detected. Therefore, compared with the case shown in FIG. 19 in the prior art, the optical input power P ₀ is related. Therefore, it is possible to detect an accurate value. Further, when the absolute value of the value to be detected is large, the amount of deviation from the optimum angle is large, so that the convergence to the optimum angle can be accelerated by performing large compensation. That is, in this embodiment, since a value proportional to the deviation (θ _i −θ ₀ ) from the optimum angle is output from the LPF 121b, the strength according to the deviation amount (for example, the control voltage according to the deviation amount). ), The convergence to the optimum angle can be accelerated.

  In addition, when the absolute value of the detected value is small, it can be determined that the angle is approaching the optimum angle, so by reducing the compensation amount, the MEMS mirror can be optimized by stable control without overshoot, undershoot, or oscillation. It is possible to converge to an angle. Furthermore, Expression (7) is a linear function with respect to the mirror angle. FIG. 3 shows the angle dependency of the low frequency component extracted in the present invention. FIG. 3 is a diagram corresponding to FIG. 18 relating to the prior art. As shown in FIG. 3, even if the angle is greatly deviated from the optimum angle, the low frequency component shown in the equation (8) is controlled so as to converge to zero, so that it can always be converged to the peak position. It becomes.

As described above, according to the first embodiment of the present invention, the logarithmic conversion unit 120 is provided, the logarithm of the signal output from the monitor circuit 10 is calculated, and then the low frequency component extraction unit 121 since so as to extract components allows the control that is independent of the optical input power P _0. In addition, since a signal proportional to the deviation amount can be obtained, it is possible to prevent the control from becoming unstable.

(B) Second embodiment

  In the first embodiment, the case where the MEMS mirror is controlled in one axial direction has been described. However, as shown in FIG. 4, the MEMS mirror can usually be controlled independently with respect to two orthogonal axes. The MEMS mirror M shown in FIG. 4 can be driven by, for example, an electrostatic force, applies a voltage to the electrode, and moves the mirror by an angle proportional to the voltage. In the example of FIG. 4, the drive axes have orthogonal X and Y axes, and each axis is driven with two electrodes (X axis drive electrode 201 and Y axis drive electrode 202) paired.

In the case of a MEMS mirror that performs two-axis control, optical coupling with a fiber collimator is expressed by a product of optical coupling efficiencies (η _x and η _y ) for each axis. Assuming that the monitored optical output power is P (θ _x , θ _y ), the optical output power can be expressed by the following equation (8).

Here, θ _n , θ _n0 , and σ _n (n = x, y) are the angle of the MEMS mirror in each axis, the optimum angle at which the optical coupling efficiency is maximized, and the output side fiber, respectively. This is the coupling radius with the collimator. In the prior art, since control is performed based on Expression (8), for example, even if a low-frequency signal is added only to the X axis in order to optimize the X axis, the optical power is equal to the X axis component and the Y axis. Since it is represented by a product of components, the monitored optical output power also depends on the optical coupling efficiency of the Y axis. Further, since the extracted low frequency component has a value proportional to the first derivative of the X-axis mirror angle (θ _x ) in Expression (8), the low frequency component similarly depends on the optical coupling efficiency of the Y axis. Value.

  Therefore, when the deviation from the optimum angle with respect to the X axis is small and the deviation from the optimum angle with respect to the Y axis is large, and conversely, the deviation from the optimum angle with respect to the X axis is large and the deviation from the optimum angle with respect to the Y axis is small. There is a possibility that a low-frequency component having the same value is extracted. In such a case, since the respective states cannot be distinguished, the prior art cannot accurately detect the optical coupling efficiency (that is, the deviation from the peak) with the fiber collimator. However, according to the present invention, even in the case of a MEMS mirror that performs biaxial control, a deviation from the optimum angle of the MEMS mirror can be detected independently for each axis.

  That is, in the case of a MEMS mirror that controls two axes, the optical monitor signal is a signal proportional to Equation (8), and therefore the output from the logarithmic converter 120 shown in FIG. 7 is expressed by Equation (9) below. Can do.

  As shown in this equation (9), as shown in the second term that depends only on the optical coupling efficiency of the X axis and the third term that depends only on the optical coupling efficiency of the Y axis, the components of the optical coupling efficiency of each axis Can be separated. For example, when a low frequency signal is added only to the X axis in order to optimize the X axis, only the second term of Equation (9) includes the low frequency component, and the third term is the same as the first term. DC component. This DC component can be removed by performing the same operation as the procedure of Equation (4) to Equation (7) in the low frequency component extraction unit 121 shown in FIG. The value depends only on the deviation from the optimum angle. That is, the DC component that depends on the optical coupling efficiency of the Y axis is multiplied by the low frequency signal cos (ωt) to become a signal having an angular frequency of ω, and can be removed by the LPF 121b. Also, when a low frequency signal is added to only the Y axis in order to optimize the Y axis, a value that depends only on the deviation from the optimum angle with respect to the Y axis can be extracted. That is, in the second embodiment of the present invention, the deviation from the optimum angle with respect to each axis can be accurately detected without depending on the optical coupling efficiency of the other axes, so that each axis can be controlled independently. It becomes.

  As a configuration example of the second embodiment, for example, when a control circuit 110 as shown in FIG. 2 is provided for each of the X axis and the Y axis and the angle in the X axis direction is adjusted, The control circuit corresponding to the X axis is operated to adjust the angle in the X axis direction. At this time, the adjustment operation is stopped in the Y-axis direction. As described above, when the low-frequency signal is superimposed on the X-axis control signal, a DC signal corresponding to the deviation from the optimum angle of only the X-axis can be obtained. Can do. The Y-axis can be adjusted to the optimum angle by the same method.

  As described above, according to the second embodiment of the present invention, the angles of the MEMS mirror in the X-axis and Y-axis directions can be adjusted independently and stably to the optimum angle.

(C) Third embodiment

  FIG. 5 is a diagram illustrating a configuration example of a control circuit 110A according to the third embodiment of the present invention. In this figure, as in the case described above, only the portion related to the “mirror angle control correction unit” for correcting the single MEMS mirror to have an optimum angle is shown in the control circuit 110A. ing. As shown in FIG. 5, in the third embodiment of the present invention, the low frequency component extraction unit 121 is replaced with a low frequency component extraction unit 121A as compared with the first embodiment. Other parts are the same as those of the first embodiment shown in FIGS. Note that an HPF (High Pass Filter) 121c is added to the low-frequency component extraction unit 121A. Other configurations are the same as those in FIG. The HPF 121c is a filter that attenuates a DC component and passes a signal having a higher frequency.

  Next, the operation of the third embodiment will be described.

When compensating for the MEMS mirror angle by adding a low-frequency signal, the low-frequency signal is used to reduce the influence on the output optical signal (the intensity of the signal light fluctuates according to the low-frequency signal). It is necessary to sufficiently reduce the amplitude (Δθ) of the MEMS mirror angle. Therefore, when the optical power (P ₀ ) is large, the direct current component included in the output from the logarithmic conversion unit 120 shown in Expression (5) may be a relatively strong power compared to the low frequency component. . In this case, the output obtained by multiplying the low-frequency signal shown in Expression (6) has other frequency components larger than the DC component, and even if only the DC component is subsequently extracted by the LPF 121b, the other frequency components May remain.

This will be specifically described. It is assumed that the signal generated by the low frequency signal generator 124 is cos (ωt), and the signal output from the logarithmic converter 120 can be expressed by the following equation (10). Here, a ₀ , a ₁ and a ₂ are coefficients corresponding to the respective frequencies.

  In the example of FIG. 2, since the output of the multiplication circuit 121a is the product of Expression (10) and cos (ωt), the following Expression (11) is obtained as the output of the multiplication circuit 121a.

As shown in equation (11), the DC component includes a coefficient a ₁ of a signal whose angular frequency is originally ω, and the component whose angular frequency is ω includes a coefficient a ₀ and a coefficient a _2. ing. Therefore, when the level of the DC signal included in the signal output from the logarithmic conversion unit 120 is high (when the coefficient a ₀ is large) and the level of the signal whose angular frequency is ω is low, the LPF 121b is set. Even after passing, a signal having an angular frequency of ω or more is included, which becomes noise.

  In particular, when the optical coupling with the fiber collimator approaches the optimum angle, the low frequency component to be extracted approaches zero, so that it is difficult to extract the low frequency component. FIG. 6 shows an example of a power spectrum output from the logarithmic converter 120 when the optical power is large in the first embodiment of the present invention. In the example of FIG. 6, since the DC component of the output signal of the logarithmic conversion unit 120 is large, not only the DC component but also other frequency components (particularly the component having an angular frequency of ω) are included in the output of the low frequency component extraction unit 121. Remains, and this becomes noise.

On the other hand, in the third embodiment, the HPF 121c attenuates the DC component from the signal output from the logarithmic conversion unit 120, and passes the other components. Thus, components of _{a 0} in the formula (10) is attenuated. As a result, in Expression (11), the value of the second term is smaller than in the first embodiment, and therefore the level of the signal whose angular frequency becomes ω after passing through the LPF 121b may be reduced. it can. This is shown in FIG. As shown in FIG. 7, the DC component is attenuated after passing through the HPF 121c. Then, after passing through the multiplication circuit 121a, the signal level of the angular frequency ω is lower than that in FIG. As a result, after passing through the LPF 121b, the level of the DC component is high, and the others are low.

  As described above, in the third embodiment of the present invention, the signal output from the logarithmic conversion unit 120 is supplied to the multiplication circuit 121a after passing through the HPF 121d. Can be extracted. Thereby, the adjustment precision of the angle of a MEMS mirror can be improved.

(D) Fourth embodiment

  FIG. 8 is a diagram illustrating a configuration example of the control circuit 110B according to the fourth embodiment of the present invention. In this figure, as in the case described above, only the part related to the “mirror angle control correction unit” for correcting the single MEMS mirror to the optimum angle in the control circuit 110B is shown. ing. As shown in FIG. 8, in the fourth embodiment of the present invention, the low frequency component extraction unit 121 is replaced with a low frequency component extraction unit 121B as compared with the first embodiment. Other parts are the same as those of the first embodiment shown in FIGS. Note that a BPF (Band Pass Filter) 121d is added to the low-frequency component extraction unit 121B. Other configurations are the same as those in FIG. The BPF 121d is a band-pass filter that passes a signal with an angular frequency ω and attenuates and outputs other signals.

  Next, the operation of the fourth embodiment will be described.

In the fourth embodiment, among the signals output from the logarithmic conversion unit 120 (see Expression (10)), signals other than those having an angular frequency of ω are attenuated by the BPF 121d. As a result, the coefficients a ₀ and a ₂ are attenuated and the coefficient a ₁ is not attenuated. As a result, the output signal of the multiplier circuit 121a (Equation (11)), and the terms other than the coefficient a ₁ is a DC component is attenuated, the output level is smaller than the first embodiment. This is shown in FIG. As shown in FIG. 9, after passing through the BPF 121d, signals other than the angular frequency ω are attenuated. Then, after passing through the multiplication circuit 121a, the signal level other than the DC component is smaller than that in FIG. As a result, after passing through the LPF 121b, only the DC component level is high, and the other frequency components are low. Thereby, the noise component is excluded.

  As described above, in the fourth embodiment of the present invention, the signal output from the logarithmic conversion unit 120 is supplied to the multiplication circuit 121a after passing through the BPF 121d. Can be extracted. Thereby, the adjustment precision of the angle of a MEMS mirror can be improved.

(E) Fifth embodiment

  FIG. 10 is a block diagram showing a configuration example of a control circuit 110C according to the fifth embodiment of the present invention. In this figure, as in the case described above, only the portion related to the “mirror angle control correction unit” for correcting the single MEMS mirror to have an optimum angle is shown in the control circuit 110C. ing. In the example shown in FIG. 10, the control circuit 110C has a control unit for each of the X axis and the Y axis, and each control unit has a low-frequency signal whose phase is shifted by π / 2. Is supplied. Specifically, the control circuit 110C includes a low frequency component extraction unit 121, a control signal generation unit 122, and a low frequency signal superimposition unit 123 for controlling the X axis, and also controls the Y axis. Low frequency component extraction unit 131, control signal generation unit 132, and low frequency signal superimposing unit 133. The low frequency component extraction unit 121 includes a multiplication circuit 121a and an LPF 121b, and the low frequency component extraction unit 131 includes a multiplication circuit 131a and an LPF 131b. The low frequency signal generation unit 124C includes a low frequency signal generation circuit 124a and a π / 2 phase shifter 124b. The configuration of each part of the control unit for each of the X axis and the Y axis is the same as in the case of FIG. The π / 2 phase shifter 124b shifts and outputs the phase of the low frequency signal output from the low frequency signal generation circuit 124a by π / 2.

  Next, the operation of the fifth embodiment will be described. The low frequency signal of the angular frequency ω output from the low frequency signal generation circuit 124a is directly supplied to the multiplication circuit 121a and the low frequency signal superimposing unit 123, and the phase is shifted by π / 2 by the π / 2 phase shifter 124b. Then, it is supplied to the multiplication circuit 131a and the low frequency signal superimposing unit 133.

Here, the angle (θ _x ) of the MEMS mirror on the X axis to which the low frequency signal output from the low frequency signal generation circuit 124a is applied, and the low frequency signal output from the π / 2 phase shifter 124b are applied. The angle (θ _y ) of the MEMS mirror on the Y axis is expressed by the following equations (12) and (13), respectively.

Note that θ _ni and Δθ _n (n = x, y) are the current mirror angle and the mirror angle amplitude based on the low-frequency signal in each axis. When Expression (12) and Expression (13) are substituted into Expression (9), an output signal from the logarithmic conversion unit 120 illustrated in FIG. 10 is represented by the following Expression (14).

  The signal output from the logarithmic conversion unit 120 includes the low frequency component of each axis independently. Here, the low-frequency signals added to the X-axis and the Y-axis have a phase shift of π / 2 from each other and are orthogonal to each other. Therefore, when cos (ωt) is multiplied by the multiplier circuit 121a, only the second term having the cos (ωt) component of Expression (14) is extracted as a DC component, and sin (ωt) is multiplied by the multiplier circuit 131a. Then, only the third term having the sin (ωt) component of Expression (14) is extracted as the DC component. Since the cos (ωt) component and the sin (ωt) component are orthogonal to each other, even if they are multiplied with each other, the result is not output as a DC component. Specifically, for example, when cos (ωt) is multiplied by the multiplication circuit 121a, the third term having a sin (ωt) component of Expression (14) is converted into a signal having only a sin (2ωt) component. Therefore, this term is not output as a DC component.

  Signals output from the multiplier circuit 121a and the multiplier circuit 131a are supplied to the LPF 121b and the LPF 131b, respectively, and components other than the DC component are attenuated and output. As a result, the output signals of the LPF 121b and the LPF 131b are as shown in Expression (15) and Expression (16), respectively.

  That is, only the low frequency component corresponding to the X axis is output from the LPF 121b, and only the low frequency component corresponding to the Y axis is output from the LPF 131b. The control signal generator 122 and the control signal generator 132 generate and output a control signal based on the signals output from the LPF 121b and the LPF 131b. As described above, the low-frequency signal superimposing unit 123 and the low-frequency signal superimposing unit 133 apply the low-frequency signal generating circuit 124a and π / to the control signals output from the control signal generating unit 122 and the control signal generating unit 132. The low frequency signals output from the two-phase shifter 124b are superimposed and output.

  As described above, according to the fifth embodiment of the present invention, the low-frequency signals orthogonal to each other are superimposed on the X-axis and Y-axis control signals and added to each axis by the multiplier circuits 121a and 131a. The low frequency components corresponding to the respective axes that have been multiplied by the frequency signals and converted into DC components by the LPFs 121b and 131b are extracted, and the angles of the X axis and the Y axis of the MEMS mirror are based on the extracted low frequency components. Since X is adjusted, the X axis and the Y axis can be adjusted independently. Moreover, since these controls can be performed simultaneously in parallel, the angle adjustment can be quickly performed as compared with the case where these are adjusted separately.

  In the fifth embodiment described above, the phase is shifted by π / 2. However, for example, the frequency may be changed between the X axis and the Y axis, or both the frequency and the phase may be changed. Good.

  In the fifth embodiment shown in FIG. 10, the control circuit of the first embodiment shown in FIG. 2 is used. However, the control circuit of the third or fourth embodiment may be used. By using these circuits, the influence of noise can be reduced.

(F) Sixth embodiment

FIG. 11 is a block diagram showing a configuration example of a control circuit 110D according to the sixth embodiment of the present invention. In this figure, as in the case described above, only the part related to the “mirror angle control correction unit” for correcting the single MEMS mirror to have an optimum angle is shown in the control circuit 110D. ing. In the example illustrated in FIG. 11, the control circuit 110 </ b> D has a low-frequency component extraction unit 121-1 to 121-n and a control signal generation unit 122-1 so that each of the MEMS mirrors M <b> 1 to Mn can be independently controlled. To 122-n, n low frequency signal superimposing units 123-1 to 123-n, and n low frequency signal generating units 124-1 to 124-n are provided. Here, the low frequency signal generating section 124-1 to 124-n, the low frequency signals of different angular frequencies ω ₁ ~ω _n respectively generated, the low-frequency component extraction unit 121-1 to 121-n and the low-frequency signal It supplies to the superimposition parts 123-1 to 123-n. Generally, for a signal having an angular frequency ω, harmonic components (ω, 2ω, 3ω,...) That are integral multiples of the basic angular frequency are generated as shown in Expression (11). The fundamental frequencies of the angular frequencies ω _{1 to} ω _n are set so that the fundamental frequency and the harmonic component do not overlap. Specifically, for example, by setting ω ₂ = 1.2 × ω ₁ , ω ₃ = 1.3 × ω ₁ ,..., The fundamental frequency and the harmonic component do not overlap. Can be set. Thus, for example, when setting ω ₂ = 2 × ω ₁ , it is possible to prevent a harmonic component (noise) twice as large as ω _{1 from} interfering with the fundamental frequency component of ω ₂ . Note that not only the frequency but also the phase may be changed. Moreover, the above is an example and it cannot be overemphasized that it may arrange | position with angular frequency intervals other than this.

  Next, the operation of the above sixth embodiment will be described.

The low frequency signal superimposing units 123-1 to 123-n are supplied from the low frequency signal generating units 124-1 to 124-n with respect to the control signals supplied from the control signal generating units 122-1 to 122-n. The low frequency signals ω _{1 to} ω _n are superposed and then supplied to the MEMS mirrors M ₁ to Mn. MEMS mirror M1~Mn, since the angular frequency contained in the supplied control signal reflection angle varies according to the low-frequency signal ω ₁ ~ω _n, the output from the optical coupler 7-1 to 7-n The optical signal is a signal whose amplitude changes in accordance with a low-frequency signal having an angular frequency of ω ₁ to ω _n .

The monitor circuit 10 converts the optical signals output from the optical couplers 7-1 to 7-n into corresponding electrical signals, adds them, and then outputs them to the control circuit 110D. Alternatively, the optical signals output from the optical couplers 7-1 to 7-n may be combined and then converted into an electrical signal for output. In the control circuit 110D, the logarithmic conversion unit 120 performs logarithmic conversion on the signal output from the monitor circuit 10, and supplies the obtained signal to each of the low-frequency component extraction units 121-1 to 121-n. To do. The low frequency component extraction units 121-1 to 121-n output the low frequency signals ω _{1 to} ω supplied from the low frequency signal generation units 124-1 to 124-n with respect to the signals output from the logarithmic conversion unit 120. After multiplying each by _n , the DC component is extracted by the LPF and output. Here, since the frequencies of the low-frequency signals are different, only the frequency components corresponding to the respective low-frequency signals ω _{1 to} ω _n are converted into DC components and extracted in each low-frequency component extraction unit. .

The control signal generators 122-1 to 122-n detect the deviation of each MEMS mirror from a desired angle in accordance with the low frequency components extracted by the low frequency component extractors 121-1 to 121-n. Then, a control signal for eliminating the deviation is generated and output to the low frequency signal superimposing units 123-1 to 123-n. The low frequency signal superimposing units 123-1 to 123-n are supplied from the low frequency signal generating units 124-1 to 124-n with respect to the control signals supplied from the control signal generating units 122-1 to 122-n. The low frequency signals ω _{1 to} ω _n are superposed and then supplied to the MEMS mirrors M ₁ to Mn.

As described above, in the sixth embodiment of the present invention, the low-frequency component extraction units 121-1 to 121-n corresponding to the number of MEMS mirrors, the control signal generation units 122-1 to 122-n, and a low-frequency signal superimposing unit 123-1 to 123-n provided, the low-frequency signal generator 124-1 to 124-n of the low-frequency signals of different angular frequencies ω ₁ ~ω _n generated by the low-frequency component extraction unit 121 -1 to 121-n and the low-frequency signal superimposing units 123-1 to 123-n are supplied, so that, for example, the adjustment operation is compared with the case where each MEMS mirror is adjusted individually (in order). Can be executed simultaneously in parallel, the time required for adjustment can be greatly reduced. In the sixth embodiment, electrical signals obtained by photoelectric conversion of the optical signals from the optical couplers 7-1 to 7-n in the monitor circuit 10 are added, and these are processed in one logarithmic conversion unit 120. Since the logarithmic converter 120 is shared, the circuit configuration can be simplified. A logarithmic conversion unit may be provided individually. Also, the optical circuit from the optical couplers 7-1 to 7-n is combined in the monitor circuit 10 and then converted into an electrical signal, which is output to the logarithmic converter 120. It is possible to use a common circuit and further simplify the circuit configuration. As described above, even when the photoelectric conversion means included in the monitor circuit 10 and the logarithmic conversion unit 120 are shared, the angular frequencies ω _{1 to} ω _n of the low frequency signal are set to different angular frequencies. The signal from each MEMS mirror can be reliably extracted.

  In the sixth embodiment shown in FIG. 11, the control circuit of the first embodiment shown in FIG. 2 is used. However, the control circuit of the third embodiment, the fourth embodiment, or the fifth embodiment is used. You may make it use. When the control circuit of the third or fourth embodiment is used, it can be made less susceptible to noise. Further, if the control circuit of the fifth embodiment is used, both the X axis and the Y axis can be controlled simultaneously.

(G) Seventh embodiment

  FIG. 12 is a block diagram showing a configuration example of the control circuit 110E according to the seventh embodiment of the present invention. In this figure, as in the case described above, only the portion related to the “mirror angle control correction unit” for correcting the single MEMS mirror to the optimum angle in the control circuit 110E is shown. ing. In the example illustrated in FIG. 12, the control circuit 110E includes a control signal generation unit 122 that is replaced with a control signal generation unit 122A, a low frequency component extraction unit 121, and a control circuit, as compared with the first embodiment illustrated in FIG. A loss value calculation unit 125 is newly inserted between the signal generation unit 122A. Here, the loss value calculation unit 125 calculates the loss value of the optical signal passing through each MEMS mirror based on the low frequency component converted into the DC component output from the LPF 121b, and supplies it to the control signal generation unit 122A. To do. The control signal generation unit 122A compares the calculated loss value supplied from the loss value calculation unit 125 with the loss target value as the target value of the loss value, and generates a control signal so that they are equal. Output. Other configurations are the same as those in FIG.

  Next, the operation of the seventh embodiment will be described.

  In general, a WDM signal may be controlled to give a certain loss value to each wavelength in order to align the optical power of each wavelength in transmission. Conventionally, when performing such constant loss control, the control voltage vs. loss value is measured in advance at the initial adjustment stage, the measurement result is stored in a memory, and the control voltage is controlled based on the stored measurement result. It was common to determine.

  FIG. 13 is a diagram showing a configuration of a conventional constant loss control circuit. The constant loss control circuit 300 acquires a control signal corresponding to the given loss target value from the loss value versus control voltage reference table 301 and outputs it as a control signal.

  However, in the conventional constant loss control, since the reference table needs to be stored in, for example, a semiconductor memory, it is necessary to increase the memory capacity, which causes an increase in cost. In addition, because the table is created based on the measurement results at the time of initial adjustment, the relationship between the control voltage and the loss value may be disrupted due to aging of the control circuit, etc., and accurate loss constant control can be performed by referring to the table. There is no possibility.

  Therefore, in the seventh embodiment, the optical coupling loss is calculated using the low frequency component extracted by the low frequency component extraction unit 121, and the constant loss control can be performed accurately. Here, since the coupling loss accompanying the optical coupling of the fiber collimator is the ratio of the optical input power to the optical output power, it is expressed by the following formula (17) when expressed in dB.

  Here, when Expression (2) is applied to Expression (17), the following Expression (18) is obtained.

  Here, the term in the parenthesis is a low frequency component represented by the equation (7). That is, if the coupling radius (σ) of the fiber collimator and the amplitude (Δθ) of the mirror angle due to the low frequency signal are known values, the optical coupling loss can be calculated from the low frequency component detected by the method of the present invention, and the table Loss constant control can be performed without reference.

  The loss value calculation unit 125 calculates the current loss value according to Expression (18) based on the low frequency component extracted by the low frequency component extraction unit 121. The control signal generation unit 122A receives the calculated loss value and the supplied (set) loss target value, and generates and outputs a new control signal.

  Here, when optimizing the optical coupling with the fiber collimator, the control is performed so that the low frequency component becomes zero (that is, the loss value is zero). In the seventh embodiment, the current loss value and the setting are set. Control is performed so that the target loss values are equal. As a result, control is performed so that the loss value of the optical signal of each wavelength is equal to the loss target value. Thereby, the optical power of each wavelength in the transmission of the WDM signal can be made uniform.

  As described above, according to the seventh embodiment of the present invention, it is possible to perform constant loss control on the optical signals of the respective wavelengths of the WDM signal. In addition, since the memory is unnecessary, the structure of the device can be simplified and the manufacturing cost can be reduced. Furthermore, it is possible to prevent the characteristics from deteriorating due to aging.

  In the seventh embodiment shown in FIG. 12, the control circuit of the first embodiment shown in FIG. 2 is used. However, the third embodiment, the fourth embodiment, the fifth embodiment, or the sixth embodiment is used. You may make it use the control circuit of a form. When the control circuit of the third or fourth embodiment is used, it can be made less susceptible to noise. Further, if the control circuit of the fifth embodiment is used, both the X axis and the Y axis can be controlled simultaneously. Furthermore, if the control circuit of 6th Embodiment is used, a some MEMS mirror can be controlled simultaneously.

(E) Modified embodiment

  In addition, said form example is an example and various deformation | transformation embodiment exists besides this. For example, in each of the above embodiments, a WDM signal is input and the WDM signal is divided into wavelengths and output from the output port. For example, a normal optical signal that is not wavelength multiplexed light is output from the input port. The light may be input, and the light may be selected and output from a desired output port.

  In each of the above embodiments, the phase of the low-frequency signal output from the logarithmic conversion unit 120 and the phase of the low-frequency signal output from the low-frequency signal generation unit 124 is not mentioned. When there is a phase difference, a delay circuit or the like for correcting the phase difference may be arranged after the logarithmic conversion unit 120 or between the low frequency signal generation unit 124 and the multiplication circuit 121a.

  In each of the above embodiments, a signal having a frequency lower than the resonance frequency of the MEMS mirror is used as the low frequency signal. However, for example, a signal having a frequency higher than the resonance frequency may be used. . Here, the low frequency signal means that the frequency is lower than that of the data signal included in the optical signal.

  In the embodiment shown in FIG. 1, an optical configuration (one-stage type mirror configuration) having only one MEMS mirror is used, but other optical configurations may be used. For example, a two-stage mirror configuration in which a MEMS mirror is further disposed between the fiber collimators 3-1 to 3-n and the fiber collimators 6-1 to 6-n and the diffraction grating 4 may be employed. In such a configuration, for example, with the second-stage MEMS mirror fixed, the corresponding first-stage MEMS mirror is adjusted to an optimum state, and then the first-stage MEMS mirror is adjusted. In a state in which is fixed in an optimal state, the corresponding second-stage MEMS mirror may be adjusted to adjust both MEMS mirrors to the optimal state. Of course, an adjustment method opposite to that described above may be used.

1 Optical signal switching device 2-1 to 2-n Optical fiber (part of input port)
3-1 to 3-n Fiber collimator (part of input port)
4 Diffraction grating 5 Condensing lens 6-1 to 6-n Fiber collimator (part of output port)
7-1 to 7-n Optical coupler 8-1 to 8-n Optical fiber (part of output port)
10 Monitor circuit (photoelectric conversion means)
11, 110, 110A, 110B, 110C, 110D, 110E Control circuit 120 Logarithmic conversion unit (logarithmic conversion means)
20, 121, 121A, 121B, 121-1 to 121-n, 131 Low frequency component extraction unit (extraction means)
20a, 121a, 131a multiplier circuit 20b, 121b, 131b LPF
121c HPF (high-pass means)
121d BPF (bandpass means)
21, 122, 122A, 122-1 to 122-n, 132 Control signal generator (generator)
22, 123, 123-1 to 123-n, 133 Low frequency signal superimposing unit (superimposing means)
23, 124, 124C, 124-1 to 124-n Low-frequency signal generator (generator)
124a Low frequency signal generation circuit 124b π / 2 phase shifter 125 Loss value calculation unit (calculation means)
200 MEMS mirror 201 X-axis drive electrode 202 Y-axis drive electrode 300 Constant loss control circuit 301 Loss value vs. control voltage reference table M1 to Mn MEMS mirror (mirror)

Claims (9)

A plurality of input ports to which optical signals are input;
A plurality of output ports for outputting optical signals;
A plurality of mirrors that reflect an optical signal input from a predetermined input port and guide the optical signal to a desired output port;
Control means for controlling a reflection angle of the mirror and selecting a propagation path of an optical signal between the input port and the output port;
The control means includes a generating means for generating a low frequency signal having a predetermined frequency;
Superimposing means for superimposing the low-frequency signal on a control signal for controlling the reflection angle of the mirror;
Photoelectric conversion means for converting a part of the optical signal output from the output port into an electrical signal;
Logarithmic conversion means for logarithmically converting the electrical signal obtained by the photoelectric conversion means;
Extraction means for extracting a signal component of the predetermined frequency included in the electrical signal output from the logarithmic conversion means;
Generating means for generating a control signal for setting the mirror at a desired reflection angle according to the extracted signal component;
An optical signal switching device. The optical signal is wavelength multiplexed light, and has a diffraction grating for entering the respective mirrors after separating the wavelength multiplexed light input from the input port into optical signals of respective wavelengths,
The control means controls the reflection angle of each mirror so that an optical signal of each wavelength is output from a desired output port.
The optical signal switching device according to claim 1, wherein: A high-pass means for attenuating a direct current component from the electrical signal output from the logarithmic conversion means and allowing a signal component of other frequency to pass through,
The optical signal switching device according to claim 1 or 2, Band pass means for passing the signal component of the predetermined frequency from the electrical signal output from the logarithmic conversion means and attenuating the signal component of the other frequency,
The optical signal switching device according to claim 1 or 2, In accordance with the signal component of the predetermined frequency included in the electrical signal output from the logarithmic conversion means, the calculation means for calculating the attenuation amount of the optical signal,
The control means controls the angle of the mirror so that the attenuation calculated by the calculation means is equal to a target attenuation.
The optical signal switching device according to claim 1, wherein the optical signal switching device is an optical signal switching device. The mirror can control the reflection angle independently about two axes orthogonal to each other,
The control means superimposes the low frequency signal on a control signal for controlling a reflection angle centered on one of the two axes, and based on the result, the reflection angle centered on the target axis is Control to the desired angle,
The optical signal switching device according to claim 1, wherein the optical signal switching device is an optical signal switching device. The mirror is capable of independently adjusting the reflection angle around two axes orthogonal to each other,
The control means superimposes a low-frequency signal having a phase difference of π / 2 on each control signal for controlling a reflection angle about the two axes, and based on these low-frequency signals, Control the reflection angle to be the desired angle,
The optical signal switching device according to claim 1, wherein the optical signal switching device is an optical signal switching device. The control means has the generation means, the superimposition means, the extraction means, and the generation means in a number corresponding to the mirror to be controlled,
The generating means generates a low frequency signal having a predetermined frequency different for each mirror,
The superimposing means superimposes the low frequency signal on a control signal to each mirror,
The logarithmic conversion means logarithmically converts the electrical signal output from the photoelectric conversion means,
Each of the extraction means extracts the signal component of the predetermined frequency of each mirror included in the electrical signal output from the logarithmic conversion means,
Each of the generating means controls each mirror to a desired reflection angle according to the extracted signal component.
The optical signal switching device according to claim 1, wherein the optical signal switching device is an optical signal switching device. A plurality of input ports to which an optical signal is input; a plurality of output ports to which an optical signal is output; and a plurality of mirrors that reflect the optical signal input from the predetermined input port and guide it to a desired output port A control means for controlling a reflection angle of the mirror and selecting a propagation path of an optical signal between the input port and the output port,
The control means generates a low frequency signal having a predetermined frequency;
A superimposing step of superimposing the low frequency signal on a control signal for controlling the reflection angle of the mirror;
A photoelectric conversion step of converting a part of the optical signal output from the output port into an electrical signal;
A logarithmic conversion step for logarithmically converting the electrical signal obtained in the photoelectric conversion step;
An extraction step of extracting a signal component of the predetermined frequency included in the electrical signal output from the logarithmic conversion step;
Generating a control signal for setting the mirror to a desired reflection angle according to the extracted signal component;
A method for controlling an optical signal switching apparatus.