JP5873054B2 - Optical transmitter and optical signal generation method - Google Patents

Description

  The present invention relates to an optical transmitter using multilevel QAM (Quadrature Amplitude Modulation) and an optical signal generation method.

As a transmission code used in an optical transmission system, multi-level QAM capable of inserting and extracting a large-capacity optical signal at a low symbol rate has attracted attention. The simplest QAM is quaternary QAM and is called QPSK (Quadrature Phase Shift Keying). In the present specification will be described with reference to QPSK as an example of multi-level QAM, but not limited to QPSK, it is applicable to the case of using all the n ^2-QAM where n is an integer. In the following description, in the drawings and mathematical formulas, when a bar (￣) is added on a character or character string such as a parameter, the character or character string such as a parameter is indicated after “￣” in the specification. This is indicated by writing.

7, in the prior art, is a diagram illustrating an exemplary configuration for generating the n ² value QAM signal using an IQ modulator 20. As shown in the figure, the IQ optical modulator 20 includes a first optical coupler 21, a first optical modulation unit 22, a second optical modulation unit 23, an optical phase shifter 24, and a second optical coupler 25. It has. CW (Continuous Wave) light input to the IQ optical modulator 20 is separated into two by the first optical coupler 21, and one of the separated CWs is input to the first optical modulator 22, and the other Is input to the second light modulator 23.

  The first light modulation unit 22 and the second light modulation unit 23 are usually configured by MZI (Mach-Zehnder Interferometer) type light modulators. The first optical modulator 22 relatively changes the optical phase and optical intensity of the CW light based on signals (Data1, １Data1) input via the input terminals In1 and In2 of the IQ optical modulator 20. . The second light modulator 23 relatively changes the optical phase and light intensity of the CW light based on the signals (Data2, ￣Data2) input to the input terminals In3 and In4 of the IQ optical modulator 20. .

  The signal (Data1, ￣Data1) is output from the first drive amplifier 14 based on the first n-value data signal. The first drive amplifier 14 generates and outputs a signal (Data1, ￣Data1) obtained by amplifying the first n-value data signal into two types of normal phase and reverse phase. The signal (Data2, ￣Data2) is output from the second drive amplifier 16 based on the second n-value data signal. Similarly to the first drive amplifier 14, the second drive amplifier 16 generates and outputs a signal (Data2, ￣Data2) obtained by amplifying the second n-value data signal into two types of normal phase and reverse phase. To do. The first n-value data signal and the second n-value data signal are n-value NRZ (Non Return-to-Zero) signals.

The signals (Data1, ￣Data1) amplified by the first drive amplifier 14 are applied to each of the two arms of the first light modulator 22 via the first drive signal electrode 221, and ± phi ₁ of the phase shift causes the CW light. The signals (Data2, ￣Data2) amplified by the second drive amplifier 16 are applied to the two arms of the second light modulator 23 via the second drive signal electrode 231 and ± φ A phase shift of ₂ is caused in the CW light. The values of the phase delays φ1 and φ2 change corresponding to n types of values that the first n-value data signal and the second n-value data signal have.

Further, the first light modulation unit 22 applies the first data bias voltage (V _bias1 , V ′ _bias1 ) to the CW light whose optical phase and light intensity are changed by the signal (Data1, ￣Data1). Based on this, a phase shift of + θ ₁ and −θ ₁ is further performed. The first data bias voltages (V _bias1 , V ′ _bias1 ) are generated in the first bias power source 42 and input via the input terminals In 5 and In 6 of the IQ optical modulator 20. The first data bias voltage (V _bias1 , V ′ _bias1 ) is applied to each of the two arms of the first optical modulation unit 22 via the first bias electrode 222, and a phase shift of ± θ ₁ Is generated in the CW light.

Further, the second optical modulator 23, the signal (Data2, ¯Data2) against CW light is changed optical phase and optical intensity at, the second data bias voltage _{(V bias2, V 'bias2)} Based on this, a phase shift of + θ ₂ and −θ ₂ is further performed. The second data bias voltage (V _bias2 , V ′ _bias2 ) is generated in the second bias power supply 52 and is input via the input terminals In 7 and In 8 of the IQ optical modulator 20. The second data bias voltage (V _bias2 , V ′ _bias2 ) is applied to each of the two arms of the second optical modulation unit 23 via the second bias electrode 232, and a phase shift of ± θ ₂ is applied to the CW. Causes light.

  Here, voltages in the signal (Data1, ￣Data1) and the signal (Data2, ￣Data2) are defined as follows. The n types of values (signal levels) of the differential signals (Data1, ￣Data1) generated by the first drive amplifier 14 are voltages V0, V1,..., Vm, −Vm,..., −V1, −V0. Is written. Note that V0> V1>...> Vm> −Vm>...> −V1> −V0. At this time, m = n / 2-1. In general, since the optical characteristics of the first light modulation unit 22 and the second light modulation unit 23 are the same, the differential signal (Data2, ￣Data2) generated by the second drive amplifier 16 has n. Similarly, the types of signal levels are V0, V1,..., Vm, -Vm, ..., -V1, -V0.

The first data bias voltages (V _bias1 , V ′ _bias1 ) are selected so that the first light modulator 22 is biased to the null point. That is, the first data is set so that the output light of the first light modulator 22 is extinguished when the differential voltage of the signal (Data1, ￣Data1) generated in the first drive amplifier 14 is zero. Bias voltages (V _bias1 , V ′ _bias1 ) are set. The second data bias voltage (V _bias2 , V ′ _bias2 ) is selected so that the second light modulator 23 is biased to the null point. Similarly to the first data bias voltage (V _bias1 , V ′ _bias1 ), when the differential voltage of the signal (Data2, ￣Data2) generated in the second drive amplifier 16 is 0, the second light The second data bias voltages (V _bias2 , V ′ _bias2 ) are set so that the output light of the modulation unit 23 is extinguished.

  The half wavelength voltage Vπ of the first light modulation unit 22 and the second light modulation unit 23 will be described. The first light modulator 22 is an MZI type modulator, and two optical waveguides are incorporated in the first light modulator 22. When the voltage Data1 and the voltage ￣Data1 applied to these two optical waveguides are both 0, the output light is extinguished as described above, and when the voltage Data1 changes to Vx and the voltage ￣Data1 changes to −Vx. When the output light reaches the maximum intensity, 2Vx is referred to as a half-wave voltage Vπ of the first light modulator 22. Even when the voltage Data1 is −Vx and the voltage ￣Data1 is Vx, the output light of the first light modulator 22 is maximized, but the optical phase of the output light in this case is different by π.

  Since the first light modulator 22 uses this property to change the phase and intensity of light, each of the signals (Data1, ￣Data1) has an amplitude of 2Vx = Vπ at the maximum, and (Data1-￣Data1) ) Is designed to have an amplitude of at most 2Vπ. The second light modulation unit 23 is also designed in the same manner as the first light modulation unit 22. 1 includes two first drive signal electrodes 221 in the first light modulation unit 22 and two second drive signal electrodes 231 in the second light modulation unit 23. The structure is such that positive and negative voltages are applied to two optical waveguides. That is, there are a total of four electrodes. This type of IQ optical modulator is called a dual drive type.

  On the other hand, the single drive IQ modulator has two drive signal electrodes. In such a configuration, an electric field is simultaneously applied to the two optical waveguides in the first optical modulation unit by the first drive signal electrode, and the two optical waveguides in the second optical modulation unit are applied by the second drive signal electrode. At the same time, an electric field is applied. Due to the anisotropy of these four optical waveguides, the same function as the dual drive type can be realized. Even when the modulator is configured in this way, n types of values (signal levels) of the data signals applied to the first drive signal electrode and the second drive signal electrode are voltages V0, V1,. , Vm, -Vm,..., -V1, -V0, and the amplitude of each drive signal is set so as not to exceed twice the half-wave voltage Vπ.

  FIG. 8 is a diagram illustrating the relationship between the electric field E1 of the output light of the first light modulation unit 22 and the voltages V0, V1,..., Vm, −Vm,. In the figure, the vertical axis represents the electric field E1 of the output light, and the horizontal axis represents the voltage difference of the signals (Data1, ￣Data1). Here, for ease of explanation, the case of a QPSK signal with n = 2 and m = 0 will be described. As shown in the figure, the electric field E1 of the output light draws a sine wave with respect to the potential of the drive signal, and the electric fields E11 and E12 of the output light generated by the voltages V0 and -V0 are symmetrical with respect to the 0 level. Located in. Further, the electric field E2 of the output light from the second light modulator 23 and the voltages V0 and -V0 are the same as the relationship shown in FIG.

Returning to FIG. 7, the description of the configuration of the IQ optical modulator 20 will be continued.
The optical phase shifter 24 performs a phase shift on the output light of the second optical modulation unit 23 based on the third bias voltage V _bias3 . The third bias voltage V _bias3 is generated by the third bias power source 64 and is input via the input terminal In9 of the IQ optical modulator 20. The third bias voltage V _bias3 is applied to the third bias electrode 241 _included in the optical phase shifter 24, and causes a phase shift (delay) in the output light of the second optical modulation unit 23. When the third bias voltage V _bias3 is an appropriate value, the phase shift amount θ ₃ is π / 2 or −π / 2, and the output light of the first light modulation unit 22 and the second light modulation unit The vector of the optical electric field with the output light of 23 is kept orthogonal. The phase shift amount θ ₃ is performed by adjusting the third bias voltage V _bias3 given to the optical phase shifter 24.

In the second optical coupler 25, the output light of the first optical modulating section 22, and the output light of the optical phase shifter 24 are multiplexed, multiplexed result is output as an optical signal of n ² values QAM. The best waveform can be obtained when the phase shift amount θ ₃ in the optical phase shifter 24 is π / 2 or −π / 2. This corresponds to ¼ of the carrier wavelength, but since the wavelength is generally on the order of a micrometer, the adjustment is very severe. Further, the optical quality of the optical QAM signals are sensitive to errors in the optical phase shifter 24, it is very important to adjust the amount of phase shift theta ₃ in the optical phase shifter 24 to an appropriate value. 7 shows a configuration in which the optical phase shifter 24 is provided at the subsequent stage of the second optical modulation unit 23, but the first optical modulation unit 22 is replaced by the subsequent stage of the second optical modulation unit 23. An optical phase shifter may be provided in the subsequent stage, or an optical phase shifter may be provided in the subsequent stage of each of the first optical modulation unit 22 and the second optical modulation unit 23.

When the output light of the first light modulation unit 22 and the output light of the optical phase shifter 24 are orthogonal, the constellation of the output light of the IQ optical modulator 20 has a lattice shape as shown in FIG. FIG. 9 is a diagram illustrating an example of a constellation of output light from the IQ optical modulator 20. FIG. 9A shows a case where θ ₃ is π / 2, and FIG. 9B shows a case where θ ₃ is −π / 2. The In-Phase component and Quadrature-Phase component in the optical electric fields E1 and E2 may be referred to as I component and Q component, respectively. Comparing the case of θ ₃ = π / 2 and the case of θ ₃ = −π / 2, the star arrangement of the constellation is inverted on the Q component side, but the demodulated signal on the Q component side in the demodulator Is inverted, the same signal can be demodulated regardless of the sign of θ ₃ . For this reason, in the case where pre-dispersion compensation is not used, even if θ ₃ = π / 2 is selected or θ ₃ = −π / 2 is selected, the demodulator inverts the demodulated signal, As a transmission system, signals can be transmitted correctly.

The optimum value of the bias voltage _{(V bias1, V 'bias1,} V bias2, V' bias2, V bias3) is known to vary over time due to a phenomenon called the bias drift. For this reason, automatic bias control is essential in commercial transceivers. If not using the pre-dispersion compensation, automatic bias control in the n ^2-QAM signal IQ modulator for generating, for example using an asymmetric bias dithering that is described in Non-Patent Document 1 and Non-Patent Document 2 Is possible.

In the above description has described n ² value QAM signal in the case where there is no pre-dispersion compensation. Next will be described the case of applying the pre-dispersion compensation n ² value QAM signal. Pre-dispersion compensation is a process of adding chromatic dispersion and polarization dispersion opposite to chromatic dispersion and polarization dispersion generated in a transmission line to an optical signal transmitted on the transmitter side. If the optical transmission line has chromatic dispersion or polarization dispersion, the quality of the received signal is degraded. If the optical signal is transmitted to the transmission line after performing pre-dispersion compensation in the transmitter in advance, the signal can be received after reducing or canceling signal quality degradation due to chromatic dispersion or polarization dispersion occurring in the transmission line. it can.

  There are several approaches to pre-dispersion compensation. One is an optical approach using a dispersion compensating fiber, and the other is an electrical approach using digital signal processing. The digital signal processing approach requires a high-speed arithmetic circuit, but has the advantage that the amount of dispersion that can be compensated is large compared to the optical approach. In the present specification, pre-dispersion compensation refers to pre-dispersion compensation of an electrical approach based on digital signal processing. In the following description, the compensation object by the pre-dispersion compensation will be described as chromatic dispersion. Chromatic dispersion is a phenomenon in which the propagation speed of an optical signal in an optical transmission line changes depending on the wavelength.

  The optical signal has an optical spectrum that spreads around the carrier frequency fc according to the modulation speed and the signal pattern. A QAM signal is an intensity modulation signal as well as an optical phase modulation, but if the same intensity and optical phase are continuous on the time axis (same sign continuous), the optical spectrum component is in the vicinity of the carrier frequency fc, If different intensities and optical phases occur alternately on the time axis (continuous different signs), the optical spectrum component is generated at a position deviating from the carrier frequency fc. This means that different wavelength components are generated depending on the signal pattern. Therefore, when the optical transmission line has chromatic dispersion, the arrival time of the optical signal changes depending on the signal pattern, and the signal quality is deteriorated.

Pre-dispersion compensation for chromatic dispersion is realized by predicting the propagation delay received on the optical transmission line by converting it to an optical phase delay ΔΦ _pre , and giving a phase shift of −ΔΦ _{pre to} the optical signal in advance at the transmitter side. The Since the magnitude and sign of the delay ΔΦ _pre depend not only on the chromatic dispersion amount of the transmission path but also on the signal pattern, real-time processing is required. FIG. 10 is a block diagram illustrating a configuration example of an optical transmitter that performs pre-dispersion compensation in a signal processing unit using a digital signal processor (DSP). In the optical transmitter shown in the figure, the signal processing unit 11 multiplies each of the first n-value data and the second n-value data by an inverse function of the transfer function of the transmission path (for example, non-patent). Reference 3). The multiplication of the inverse function of the transfer function is an operation for delaying the optical phase of the optical signal to be transmitted by −ΔΦ _pre . The first n-value data and the second n-value data to which the delay of ΔΦ _pre is added are the first digital-analog converter (DAC) 12 and the second digital-analog converter 13. Is converted into an analog signal and input to the first drive amplifier 14 and the second drive amplifier 16. As a result, the signals (Data1, ￣Data1) and the signals (Data2, ２Data2) input to the IQ optical modulator 20 become drive signals subjected to pre-dispersion compensation of chromatic dispersion.

If the arithmetic processing performed by the signal processing unit 11 is simplified in the phase space, the processing shown in FIG. 11 is obtained. FIG. 11 is a diagram illustrating an example of pre-dispersion compensation. Here, QPSK is used for modulation, and θ ₃ = π / 2. As described above, the magnitude and sign of ΔΦ _pre depend on the chromatic dispersion amount of the transmission line and the signal pattern, and thus change from moment to moment, and are extremely complicated. It illustrates a case of delaying the optical phase of the symbol a + .DELTA..PHI _pre only in 9. As is apparent from FIG. 11, this calculation process is a process of multiplying the electric field E11 and the electric field E21 by a rotation matrix. If θ ₃ = π / 2, this arithmetic processing may be performed by conversion as shown in the following equation.
_{E11 ← E11 × cos (ΔΦ pre} ) + E21 × sin (ΔΦ pre)
_{E21 ← E11 × cos (ΔΦ pre} ) -E21 × sin (ΔΦ pre)

H. Kawakami, E. Yoshida, and Y. Miyamoto, "Auto Bias Control Technique Based on Asymmetric Bias Dithering for Optical QPSK Modulation," Journal of Lightwave Technology, Vol. 30, No. 7, April 1, 2012 H. Kawakami, T. Kobayashi, E. Yoshida, and Y. Miyamoto, "Auto bias control technique for optical 16-QAM transmitter with asymmetric bias dithering," Optics Express, Vol.19, Issue26, pp.B308-B312, 2011 T. Kobayashi, S. Kametani, Y. Konishi, T. Sugihara, S. Hirano, K. Tsutsumi, K. Yamagishi, T. Ichikawa, S. Inoue, K. Kubo, Y. Takahashi, K. Goto, K. Uto, T. Yoshida, K. Sawada, H. Bessho, K. Koguchi, K. Shimizu, and T. Mizuochi, "43Gb / s DQPSK Transmission over Fiber with 2450 ps / nm of Dispersion using Electronic Pre-equalization," ECOC 2010, 19-23 September 2010

Here, let us consider a case where the sign of θ ₃ is changed for some reason and θ ₃ = −π / 2 (3π / 2). FIG. 12 is a diagram illustrating an example when θ ₃ in FIG. 11 is changed from π / 2 to −π / 2. As shown in FIG. 12, the optical phase advances by ΔΦ _{pre in the} opposite direction to the originally intended direction. As described above, since the sign of the delay amount ΔΦ _pre is an amount determined depending on the chromatic dispersion amount of the transmission line, in this case, correct pre-chromatic dispersion compensation cannot be performed.

From the above, in the bias control in the optical transmitter for generating the n ² value QAM signal with the pre-dispersion compensation, an optical phase difference between the output light of the first optical modulating section 22 and the second optical modulating section 23 In addition to controlling the absolute value of θ ₃ to π / 2, a function that can correctly select the sign of θ ₃ is required.

  The present invention has been made in view of such circumstances, and an object of the present invention is to control the amount of phase shift when generating a multi-level QAM signal combined with electrical pre-dispersion compensation by digital signal processing. It is an object to provide an optical transmitter and an optical signal generation method capable of performing the above.

  According to one embodiment of the present invention, the first n-value data to be transmitted is transmitted to one continuous wave light among the first optical coupler that separates continuous wave light and the continuous wave light separated by the first optical coupler. Based on the first optical modulation unit that changes the optical phase and the optical intensity, and the second n-value data to be transmitted with respect to the other continuous wave light among the continuous wave light separated in the first optical coupler. And a phase that shifts the phase of the output light of one of the second light modulation unit that changes the light phase and the light intensity, and the first light modulation unit or the second light modulation unit. A shifter, a second optical coupler that combines the output light of the other of the first light modulation unit and the second light modulation unit, and the output light of the phase shifter; When the n-value data is at a zero level, the output light of the first light modulator is quenched. A first bias power source for generating a first data bias voltage, and a second data bias voltage for quenching the output light of the second optical modulation unit when the second n-value data is at a zero level. A second bias power source, an optical power monitor that outputs the optical power of the output light of the second optical coupler as an electrical signal, and a first dither having a first frequency as a first data bias voltage First dither signal applying means for superimposing a signal; and second dither signal applying means for superimposing a second dither signal whose phase is orthogonal to the first dither signal on the second data bias voltage; The electrical signal output from the optical power monitor is synchronously detected using a second frequency that is an integral multiple of the first frequency, and the phase shift amount in the phase shifter is corrected so that the synchronous detection result becomes zero. Phase amount correction means, wherein the phase amount correction means increases the phase shift amount when the synchronous detection result is negative, and increases the phase shift amount when the synchronous detection result is positive. One of a positive slope mode for decreasing and a negative slope mode for increasing the phase shift amount when the synchronous detection result is positive and decreasing the phase shift amount when the synchronous detection result is negative. The optical transmitter is selected in advance and corrects a phase shift amount in the phase shifter based on the selected slope mode and the synchronous detection result.

  According to another aspect of the present invention, the optical transmitter further includes a changeover switch that enables switching between the positive slope mode and the negative slope mode as needed.

  Further, according to one embodiment of the present invention, in the optical transmitter described above, pre-dispersion compensation according to chromatic dispersion or polarization dispersion occurring in a transmission path is applied to the first n-value data and the second n-value data. And a first digital output unit that converts the first n-value data subjected to pre-dispersion compensation in the signal processing unit into an analog signal and outputs the analog signal to the first optical modulation unit. An analog conversion unit, and a second digital-analog conversion unit that converts the second n-value data subjected to pre-dispersion compensation in the signal processing unit into an analog signal and outputs the analog signal to the second optical modulation unit And further comprising.

  According to another aspect of the present invention, in the above optical transmitter, the signal processing unit performs low-pass filter processing on the first n-value data and the second n-value data. .

  According to another aspect of the present invention, in the above optical transmitter, the electrical signal output from the optical power monitor is synchronously detected using the first dither signal, and the synchronous detection result is 0. A first bias correction means for correcting the first data bias voltage generated by the first bias power supply, and an electrical signal output from the optical power monitor, using the second dither signal, and synchronously detecting, And a second bias correction unit that corrects the second data bias voltage generated by the second bias power supply so that the synchronous detection result becomes zero.

  According to another aspect of the present invention, in the optical transmitter described above, at least one of the phase amount correction unit, the first bias correction unit, and the second bias correction unit generates a synchronous detection result. On the other hand, correction based on the result of adding a predetermined offset value is performed.

  According to one embodiment of the present invention, a first optical coupler that separates continuous wave light and a first n value that is transmitted with respect to one continuous wave light among the continuous wave lights separated by the first optical coupler. Based on the data, the first optical modulation unit that changes the optical phase and the light intensity, and the second n value to be transmitted with respect to the other continuous wave light among the continuous wave light separated in the first optical coupler Based on the data, a phase shift is performed on the output light of the second light modulation unit that changes the optical phase and the light intensity, and either the first light modulation unit or the second light modulation unit. A phase shifter to be performed; a second optical coupler that combines the output light of the other of the first light modulation unit and the second light modulation unit and the output light of the phase shifter; When the n-value data of 1 is zero level, the output light of the first light modulation unit is turned off. A first bias power source that generates a first data bias voltage to be generated, and a second data bias voltage that causes the output light of the second optical modulation unit to be extinguished when the second n-value data is at a zero level. A second bias power source that is generated, an optical power monitor that outputs the optical power of the output light of the second optical coupler as an electrical signal, and a first having a predetermined first frequency for the first data bias voltage First dither signal applying means for superimposing a dither signal; and second dither signal applying means for superimposing a second dither signal whose phase is orthogonal to the first dither signal on the second data bias voltage. The electric signal output from the optical power monitor is synchronously detected using a second frequency that is an integral multiple of the first frequency, and the phase shift amount in the phase shifter is set so that the synchronous detection result becomes zero. A phase amount correcting means for correcting, a positive slope mode for increasing the phase shift amount when the synchronous detection result is negative, and decreasing the phase shift amount when the synchronous detection result is positive; and the synchronization Optical transmission comprising switching means for selecting one of a negative slope mode that increases the phase shift amount when the detection result is positive and decreases the phase shift amount when the synchronous detection result is negative An optical signal generation method in an optical device comprising: a first step for determining a magnitude and a sign of a control signal based on the synchronous detection result and the slope mode selected by the switching means; and the control signal to the phase shifter And a second step of correcting the phase shift amount by feedback. It is.

  According to the present invention, the intensity modulation component (synchronous detection) detected by synchronous detection when the phase difference between the two output lights combined in the second optical coupler deviates from π / 2 or −π / 2. In addition to controlling the absolute value of the phase shift amount to π / 2 by correcting the phase shift amount based on the result), the phase shift amount is properly adjusted so that electrical pre-dispersion compensation by digital signal processing is correctly performed. Can be defined.

It is a figure which shows the transition of the optical electric field E1 at the time of superimposing a dither signal on the 1st data bias voltage. It is a 1st figure which shows the periodic distortion by the asymmetric bias dithering which arises in the constellation of QPSK. It is a 2nd figure which shows the periodic distortion by the asymmetrical bias dithering which arises in the constellation of QPSK. It is a graph which shows a result when performing synchronous detection using sin (2ωd × t) as a reference clock. It is a block diagram which shows the structure of the optical transmitter in 1st Embodiment based on this invention. It is a block diagram which shows the structure of the optical transmitter in 2nd Embodiment. In the prior art, it is a diagram illustrating an exemplary configuration for generating the n ² value QAM signal using an IQ modulator 20. The electric field E1 of the output light of the first light modulator 22 and the voltages V0, V1,..., Vm, -Vm, ..., _-V1 , _-V0 , and the first data bias voltage ( _Vbias1 , _V'bias1 ). It is a figure which shows the relationship. 3 is a diagram illustrating an example of a constellation of output light from an IQ optical modulator 20. FIG. It is a block diagram which shows the structural example of the optical transmitter which performs pre-dispersion compensation in the signal processing part using a digital signal processor (Digital Signal Processor: DSP). It is a figure which shows an example of pre-dispersion compensation. FIG. 12 is a diagram illustrating an example when θ ₃ in FIG. 11 is changed from π / 2 to −π / 2.

Before describing the configuration and the like of the optical transmitter in the embodiment according to the present invention, a state where pre-dispersion compensation is not used will be described again. Here, the first data bias voltage _{(V bias1, V 'bias1)} and the second data bias voltage _{(V bias2, V' bias2)} is asymmetrical bias that is described in Non-Patent Document 1 and Non-Patent Document 2 It is assumed that the first optical modulation unit 22 and the second optical modulation unit 23 (FIGS. 7 and 10) are correctly biased to the null point because the optimum values have already been maintained by dithering.

  In asymmetric bias dithering, at least one of the voltages V0, V1,..., Vm is set smaller than Vπ, and a low-speed dither signal whose phase is orthogonal to the first data bias voltage and the second data bias voltage. Superimpose. In the case of QPSK with n = 2 and m = 0, the optical electric field E1 is a symbol of a double circle (◎) or a white circle (◯) with the two filled black circles (●) shown in FIG. 1 as the center. Transition periodically. FIG. 1 is a diagram illustrating a transition of the optical electric field E1 when a dither signal is superimposed on the first data bias voltage. In the figure, the vertical axis represents the electric field E1 of the output light, and the horizontal axis represents the voltage difference of the signals (Data1, ￣Data1).

  Here, it should be noted that when the dither signal is not applied (black circle (●)), the absolute values of the electric field E11 and the electric field E12 are equal, but the dither signal is positive (double circle (（) )), | E11 |> | E12 |, and when the dither signal is negative (white circle (◯)), | E11 | <| E12 |. As a result, the constellation is periodically distorted slightly asymmetrically with respect to the origin.

The dither signal applied to the first data bias voltage is represented by cos (ωd × t), and the dither signal applied to the second data bias voltage is represented by sin (ωd × t). FIG. 2 shows periodic distortion caused by asymmetric bias dithering, which occurs in the QPSK constellation. FIG. 2 is a diagram showing periodic distortion due to asymmetric bias dithering that occurs in the constellation of QPSK. Here, how the optical power changes due to this distortion will be described. Here, the “optical power” is a value obtained by averaging at a period that is much longer than the symbol period of the signal (˜100 picoseconds) and shorter than the dithering period (˜1 millisecond). When theta ₃ is [pi / 2, the symbols that protrudes whatever value is .omega.d × t does not occur.

However, when θ is smaller than π / 2 (the row of θ ₃ = 45 °), when ωd × t = π / 4 and ωd × t = 5π / 4, the most dissimilar symbol (blacked out) An asterisk (★) is generated. A black star (★) and a symbol closest to the origin (white star (☆)) appear at the opposite position. Since the optical power is proportional to the square of the optical electric field vector, the deviation between the two is not canceled out, and the total optical power is maximized. Therefore, as described in Non-Patent Document 1, an intensity modulation component having a period 2ωd that is twice the frequency of the dither signal is superimposed on the optical power of the entire optical QPSK.

On the other hand, when θ is larger than π / 2 (θ ₃ = 135 ° column), an intensity modulation component with a period 2ωd that is twice the frequency of the dither signal is superimposed on the optical power of the entire optical QPSK. The peaks appear at ωd × t = 3π / 4 and ωd × t = 7π / 4.

theta ₃ is increased, showing a case where it exceeds π in FIG. FIG. 3 is a diagram illustrating periodic distortion caused by asymmetric bias dithering in the QPSK constellation. Since the axis of E2 is inverted vertically, the influence of the dither signal sin (ωd × t) is also inverted. If θ ₃ = 3π / 2, but does not occur symbols protruding whatever value is .omega.d × t, when theta ₃ is less than _{3π / 2 (θ 3 = 225} ° of the column), .omega.d The optical power becomes maximum at xt = 3π / 4 and ωd × t = 7π / 4. In addition, when θ ₃ is larger than 3π / 2 (a row of θ ₃ = 315 °), the optical power becomes maximum at ωd × t = π / 4 and ωd × t = 5π / 4.

From the above description, when the pre-dispersion compensation is not used and θ ₃ is not a right angle (π / 2 or 3π / 2), the intensity modulation component of 2ωd is superimposed on the optical power of the optical QPSK. Is shown. Here, it is assumed that the intensity modulation component is synchronously detected using a sin (2ωd × t) sine wave as a reference clock. Since sin (2ωd × t) takes the maximum value 1 at ωd × t = π / 4 and ωd × t = 5π / 4, the synchronous detection result can be expressed as a function of θ ₃ as shown in FIG. FIG. 4 is a graph showing the results when synchronous detection is performed using sin (2ωd × t) as a reference clock. In the figure, the ordinate indicates the synchronous detection result, and the horizontal axis shows the theta _3.

As shown in FIG. 4, the synchronous detection result becomes 0 when θ ₃ is π / 2 (= 90 °) or 3π / 2 (= 270 °). However, the slope of the synchronous detection result with respect to θ ₃ is reversed. Since 3π / 2 and −π / 2 are equivalent, by searching for the zero point after limiting the slope (tilt) of the synchronous detection result, θ ₃ = π / 2 or θ ₃ = −π / 2 is obtained. It becomes possible to select correctly.

In FIG. 4, the feedback control for searching for the zero point so that the slope near the zero point becomes lower right is defined as negative slope mode, and the feedback control for searching for the zero point so that the slope near the zero point rises to the right is positive slope mode. It is defined as In the negative slope mode, it converges to θ ₃ = + π / 2, and in the positive slope mode, it converges to θ ₃ = −π / 2. This correspondence also depends on the circuit configuration and may be reversed depending on the arrangement of the electrodes to which the dither signal is applied and the structure of the optical power monitor. It can also be changed by the pre-dispersion compensation algorithm. However, once definite circuit configuration and algorithm, the sign of the polarity and theta ₃ slope mode to be selected is uniquely determined.

  Before performing feedback control, the correct slope mode must be selected in advance. This is realized by determining the relative relationship between the sign of the feedback signal and the magnitude of the error signal in feedback control. If the device configuration is unchanged, the slope mode may be fixed. However, when a change in the pre-dispersion compensation algorithm is predicted, it is desirable that the slope mode can be switched with the selection switch so that the change can be accommodated.

In the above description, for the sake of simplicity, the description has been made in a state where pre-dispersion compensation is not used. Next, a method for selecting a sign of θ _{3 in} a state where pre-dispersion compensation is performed in the embodiment according to the present invention will be described. The change in the angle of the symbol by the pre-dispersion compensation shown in FIG. 11 is a conversion that keeps the distance from the origin to the symbol constant. Since the optical power is proportional to the square of the magnitude of the vector of the optical electric field, the conversion shown in FIG. 11 does not change the optical power. Therefore, whether or not θ ₃ is + π / 2 or −π / 2 can be correctly determined by applying the method shown in FIG. 4 as it is regardless of the presence or absence of pre-dispersion compensation. Description up to this point has been the optical QPSK signal as an example, it can be treated similarly for ordinary light n ² value QAM signal.

<First Embodiment>
FIG. 5 is a block diagram showing a configuration of the optical transmitter according to the first embodiment of the present invention. In the figure, the same reference numerals are given to the same functional units as those described in FIGS. 7 and 10. Further, the description of the functional units described in FIG. 7 or 10 is omitted. In FIG. 5, the description of the configuration in the IQ optical modulator 20 is omitted, input terminals In1 to In9 are shown, and the connection relationship between the IQ optical modulator 20 and other configurations is shown. In the optical transmitter shown in the figure, signals (Data1, ￣Data1) are input to the input terminals In1 and In2 of the dual drive IQ optical modulator 20, and signals (Data3, ￣Data1) are input to the input terminals In3 and In4 of the IQ optical modulator 20, respectively. Data2 and ￣Data2) are input.

  The first n-value data that is the first n-value data that has been subjected to the pre-dispersion compensation in the signal processing unit 11 and converted into an analog signal is input to the first drive amplifier 14. In the first drive amplifier 14, a differential signal (Data1, ￣Data1) of the input first n-value data is generated, and the amplitude of the differential signal is adjusted by the first amplitude adjustment unit 15. The That is, the first drive amplifier 14 generates a differential signal (Data1, ￣Data1) whose amplitude is adjusted by the first amplitude adjustment unit 15 from the first n-value data inputted thereto, and IQ Input to the optical modulator 20.

  Similarly, second n-value data that has been subjected to pre-dispersion compensation in the signal processing unit 11 and converted into an analog signal is also input to the second drive amplifier 16. The second drive amplifier 16 generates a differential signal (Data2, ２Data2) of the input second n-value data, and the amplitude of the differential signal is adjusted by the second amplitude adjustment unit 17. The That is, the second drive amplifier 16 generates a differential signal (Data2, ￣Data2) whose amplitude is adjusted by the second amplitude adjustment unit 17 from the inputted second n-value data, and IQ Input to the optical modulator 20. The first amplitude adjusting unit 15 and the second amplitude adjusting unit 17 adjust the amplitude so that at least one of the aforementioned voltages V0, V1,..., Vm is smaller than Vπ.

In the present embodiment, the voltage output from the first bias power supply 42 is differentially amplified by the first differential amplifier 44. The first data bias voltage ± V _bias1 differentially amplified by the first differential amplifier 44 is input to the input terminals In5 and In6 of the IQ optical modulator 20 and applied to the first bias electrode 222. The first data bias voltage ± V _bias1 is applied to two arms of the first light modulation unit 22 in the IQ light modulator 20. The two arms receive a differential bias voltage by the first data bias voltage ± V _bias1 .

The voltage output from the first bias power source 42 is subjected to dithering in which a sine wave signal having a frequency ωd output from the first oscillator 41 is added by the first adder 43 to obtain a first difference. Input to the dynamic amplifier 44. The amplitude of dithering to the first data bias voltage ± _{V bias1} is smaller than the baud rate of ^{n 2-QAM} signal, to at most kHz order.

The voltage output from the second bias power supply 52 is differentially amplified by the second differential amplifier 54. Second data bias voltage ± _{V bias2} that are differentially amplified in the second differential amplifier 54 is inputted to the input terminal In7 and In8 the IQ modulator 20 is applied to the second bias electrode 232. The second data bias voltage ± V _bias2 is applied to two arms of the second optical modulation unit 23 in the IQ optical modulator 20. The two arms _receive a differential bias voltage by the second data bias voltage ± V _bias2 .

The voltage output from the second bias power source 52 is subjected to dithering in which a sine wave signal having a frequency ωd output from the second oscillator 51 is added by the second adder 53, and the second difference is applied. Input to the dynamic amplifier 54. The amplitude of the dithering for the second data bias voltage ± _{V bias2} is smaller than the baud rate of ^{n 2-QAM} signal, to at most kHz order.

In the present embodiment, the output of the first oscillator 41 is cos (ωd × t), and the output of the second oscillator 51 is sin (ωd × t). That is, the sine wave signal output from the first oscillator 41 and the sine wave signal output from the second oscillator 51 are orthogonal to each other. Note that t is time. Therefore, in the optical transmitter in FIG. 5, it takes dithering synchronized with a phase shift theta ₁ to cos (ωd × t), the phase shift amount theta ₂ takes dithering synchronized with sin (ωd × t) Yes.

In the IQ optical modulator 20, the optical phase shifter 24 performs phase shift on the output light of the second optical modulation unit 23 in accordance with the third bias voltage V _bias3 output from the third bias power supply 64. Is done. For the third bias voltage _{V bias3,} unlike the first data bias voltage ± _{V bias1} and the second data bias voltage ± _{V bias2,} not subjected to dithering. N ^2-QAM signal output from the IQ modulator 20 is branched by the third optical coupler 31, one is transmitted, the other is input to the optical Pawamonita 32.

  The optical power monitor 32 measures the optical power in the optical signal branched by the third optical coupler 31 and inputs the measurement result to the third synchronous detector 62 as an electric signal. The band that can be monitored (measured) by the optical power monitor 32 may be such that it can follow twice the dithering frequency ωd. The third synchronous detector 62 performs synchronous detection on the signal input from the optical power monitor 32 using the sin (2ωd × t) sine wave signal output from the third oscillator 61 as a reference clock. The third synchronous detector 62 inputs the synchronous detection result to the third loop gain adjustment circuit 63.

  The third loop gain adjustment circuit 63 determines the magnitude of the feedback signal using PID control, and based on the slope mode selected from the positive slope mode and the negative slope mode and the synchronous detection result. Determine the sign of. The slope mode is selected in advance before performing feedback control. Note that a switch that enables switching between the positive slope mode and the negative slope mode at any time may be mounted in the third loop gain adjustment circuit 63 so as to cope with a change in the pre-dispersion compensation algorithm.

As shown in FIG. 4, when the negative slope mode is selected, the third bias voltage V _bias3 is changed so that the phase shift amount θ ₃ increases when the synchronous detection result is positive, and the synchronous detection result The third loop gain adjustment circuit 63 controls the third bias power supply 64 to change the third bias voltage V _bias3 so as to decrease the phase shift amount θ ₃ when is negative. When the positive slope mode is selected, the third loop gain adjustment circuit 63 performs the control opposite to the above-described control on the third bias power supply 64.

Further, in the optical transmitter according to the present embodiment, electrical pre-conditioning by digital signal processing can be performed without performing complicated processing such as defining two inverse functions of the transfer function according to whether the phase shift amount θ ₃ is positive or negative. dispersion compensation process corrects the amount of phase shift theta ₃ to properly added n ² value QAM signal is generated, the output light of the output light and the optical phase shifter 24 of the first optical modulating section 22 The codes can be orthogonalized with the correct selection.

<Second Embodiment>
FIG. 6 is a block diagram illustrating a configuration of the optical transmitter according to the second embodiment. In the figure, the same reference numerals are given to the same functional units as those described in FIGS. Further, description of the functional units described in FIGS. 5, 7, and 10 is omitted. Similarly to FIG. 5, in FIG. 6, the description of the configuration in the IQ optical modulator 20 is omitted, the input terminals In1 to In9 are shown, and the connection relationship between the IQ optical modulator 20 and other configurations is shown. The optical transmitter according to the present embodiment includes a first synchronous detector 45 and a first loop gain adjustment circuit 46, and a second synchronous detector 55 and a second loop gain adjustment circuit 56. This is different from the optical transmitter (FIG. 5) in the first embodiment.

The optical transmitter according to the present embodiment uses the first synchronous detector 45 and the first loop gain adjustment circuit 46, the second synchronous detector 55 and the second loop gain adjustment circuit 56, and an optical power monitor. Feedback control using 32 outputs is performed on the first bias power source 42 and the second bias power source 52. Thus, to maintain the first and second data bias voltage (± _{V bias1} and ± _{V bias2)} to the optimum value is varied depending on the bias drift.

The first synchronous detector 45 performs synchronous detection on the signal input from the optical power monitor 32 using the sine wave signal (cos (ωd × t)) output from the first oscillator 41 as a reference clock. The first synchronous detector 45 inputs the synchronous detection result to the first loop gain adjustment circuit 46. The first loop gain adjustment circuit 46 controls the first data bias voltage V _bias1 output from the first bias power supply 42 so that the input synchronous detection result becomes 0 using PID control or the like. .

The second synchronous detector 55 performs synchronous detection on the signal input from the optical power monitor 32 using the sine wave signal (sin (ωd × t)) output from the second oscillator 51 as a reference clock. The second synchronous detector 55 inputs the synchronous detection result to the second loop gain adjustment circuit 56. The second loop gain adjustment circuit 56 uses the PID control or the like to control the second data bias voltage V _bias2 output from the second bias power supply 52 so that the input synchronous detection result becomes zero. .

In the optical transmitter in the present embodiment, by changing the first and second data bias voltage according to the bias drift occurs in the first and second data bias voltage _{(V bias1,} V _bias2), is output it is possible to improve the quality of the n ² value QAM signal. In the present embodiment, the configuration using three synchronous detectors (first synchronous detector 45, second synchronous detector 55, and third synchronous detector 62) has been described. A configuration may be adopted in which bias drift is detected and fed back using a single synchronous detector by time sharing.

In the optical transmitters in the above-described embodiments, the first data bias voltage _Vbias1 and the second data bias voltage _Vbias2 are differentially driven, but may be single-phase driven by grounding one of them. Good. In this case, for example, the input terminals In6 and In8 of the IQ optical modulator 20 are grounded.
In the optical transmitters in the above-described embodiments, the drive signal is described as dual drive, but single drive may be performed using an X-cut modulator.

  In the optical transmitter in each of the above-described embodiments, the case where the optical power monitor 32 is provided outside the IQ optical modulator 20 has been described. However, the IQ optical modulator may include an optical power monitor. For example, some commercially available light modulators have a built-in optical power monitor, which may be used. However, the optical power monitor built in the optical modulator may have an error called peak shift. Since the modulator is an interference system, the output power of the optical modulator changes periodically when each bias is changed, and the output of the optical power monitor changes periodically in synchronization with this, but the peak shift described above is ignored. If this is not possible, the peaks of these two periodic variations will be slightly out of order. Therefore, when bias control is performed using such an optical power monitor, the control loop converges to a bias value slightly deviated from the optimum value. In order to avoid this, each of the first to third loop gain adjustment circuits may have an offset function, and a feedback signal may be generated after adding a constant for fine adjustment to the synchronous detection result.

In addition, in order to improve the utilization efficiency of the optical frequency, the signal processing unit 11 is provided with a digital LPF (Low Pass Filter) function to remove noise and harmonic components contained in the first and second n-value data. The low pass filter processing may be performed. At this time, for example, a Nyquist filter may be used as the digital LPF.
In the optical transmitter in each of the above-described embodiments, the third oscillator 61 performs synchronous detection using a sine wave signal (sin (2ωd × t)) having a frequency twice that of the dither signal. As described above, synchronous detection may be performed using a sine wave signal of integer multiple instead of double.

  You may make it implement | achieve the optical transmitter in each embodiment mentioned above combining an IQ optical modulator and a computer. In that case, a program for realizing this function may be recorded on a computer-readable recording medium, and the program recorded on this recording medium may be read into a computer system and executed. Here, the “computer system” includes an OS and hardware such as peripheral devices. The “computer-readable recording medium” refers to a storage device such as a flexible medium, a magneto-optical disk, a portable medium such as a ROM and a CD-ROM, and a hard disk incorporated in a computer system. Further, the “computer-readable recording medium” is a program that dynamically holds a program for a short time, like a communication line when a program is transmitted via a network such as the Internet or a communication line such as a telephone line. In this case, a volatile memory inside a computer system serving as a server or a client in that case may be included and a program held for a certain period of time. Further, the program may be for realizing a part of the functions described above, and may be a program capable of realizing the functions described above in combination with a program already recorded in a computer system. It may be realized using hardware such as PLD (Programmable Logic Device) or FPGA (Field Programmable Gate Array).

  The embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and includes designs and the like that do not depart from the gist of the present invention.

DESCRIPTION OF SYMBOLS 11 ... Signal processing part 12 ... 1st digital-analog converter 13 ... 2nd digital-analog converter 14 ... 1st drive amplifier 15 ... 1st amplitude adjustment part 16 ... 2nd drive amplifier 17 ... 1st 2. Amplitude adjustment unit 20... IQ light modulator 21... First optical coupler 22... First light modulation unit 221... First drive signal electrode 222 ... First bias electrode 23. 231 ... Second drive signal electrode 232 ... Second bias electrode 24 ... Optical phase shifter 241 ... Third bias electrode 25 ... Second optical coupler 31 ... Third optical coupler 32 ... Optical power monitor 41 ... First Oscillator 42... First bias power supply 43... First adder 44... First differential amplifier 45 .. first synchronous detector 46 .. first loop gain adjustment circuit 51. 2 bias power supply 5 DESCRIPTION OF SYMBOLS 3 ... 2nd adder 54 ... 2nd differential amplifier 55 ... 2nd synchronous detector 56 ... 2nd loop gain adjustment circuit 61 ... 3rd oscillator 62 ... 3rd synchronous detector 63 ... 3rd Loop gain adjustment circuit 64 ... third bias power supply

Claims (6)

A first optical coupler for separating continuous wave light;
A first light modulation unit that changes an optical phase and light intensity based on first n-value data to be transmitted with respect to one continuous wave light among the continuous wave lights separated in the first optical coupler;
A second optical modulation unit that changes an optical phase and an optical intensity based on second n-value data to be transmitted with respect to another continuous wave light among the continuous wave lights separated in the first optical coupler;
A phase shifter that performs a phase shift on the output light of either the first light modulation unit or the second light modulation unit;
A second optical coupler that combines the output light of the other of the first light modulation unit or the second light modulation unit and the output light of the phase shifter;
A first bias power supply for generating a first data bias voltage for quenching the output light of the first light modulation unit when the first n-value data is at a zero level;
A second bias power source for generating a second data bias voltage for quenching the output light of the second optical modulation unit when the second n-value data is at a zero level;
An optical power monitor that outputs the optical power of the output light of the second optical coupler as an electrical signal;
First dither signal applying means for superimposing a first dither signal having a predetermined first frequency on the first data bias voltage;
Second dither signal applying means for superimposing a second dither signal having the same frequency as the first frequency and having a phase orthogonal to the first dither signal on the second data bias voltage When,
The electrical signal output from the optical power monitor is synchronously detected using a second frequency that is an integer multiple of the first frequency, and the integer is an even number equal to or greater than 2, so that the synchronous detection result is zero. Phase amount correction means for correcting a phase shift amount in the phase shifter;
When the synchronous detection result is negative, the phase shift amount is increased, and when the synchronous detection result is positive, the positive slope mode for decreasing the phase shift amount, and when the synchronous detection result is positive A change-over switch that increases the phase shift amount and enables switching to a negative slope mode that reduces the phase shift amount when the synchronous detection result is negative;
With
The phase amount correction means includes
Either the positive slope mode or the negative slope mode is selected in advance, and the phase shift amount in the phase shifter is corrected based on the selected slope mode and the synchronous detection result. Optical transmitter. The optical transmitter according to claim 1,
A signal processor that performs pre-dispersion compensation on the first n-value data and the second n-value data according to chromatic dispersion or polarization dispersion generated in the transmission line;
A first digital-analog converter that converts the first n-value data subjected to pre-dispersion compensation in the signal processor into an analog signal and outputs the analog signal to the first optical modulator;
A second digital-analog converter that converts the second n-value data subjected to pre-dispersion compensation in the signal processor into an analog signal and outputs the analog signal to the second optical modulator. An optical transmitter characterized by. The optical transmitter according to claim 2, wherein
The signal processing unit
An optical transmitter, wherein low pass filter processing is performed on the first n-value data and the second n-value data. The optical transmitter according to any one of claims 1 to 3,
The electrical signal output from the optical power monitor is synchronously detected using the first dither signal, and the first data bias voltage generated by the first bias power supply is set so that the synchronous detection result becomes zero. First bias correction means for correcting;
The electrical signal output from the optical power monitor is synchronously detected using the second dither signal, and the second data bias voltage generated by the second bias power supply is set so that the synchronous detection result becomes zero. And a second bias correcting means for correcting the optical transmitter. The optical transmitter according to claim 4, wherein
At least one of the phase amount correction means, the first bias correction means, and the second bias correction means is:
An optical transmitter characterized by performing correction based on a result obtained by adding a predetermined offset value to a synchronous detection result. Based on the first n-value data to be transmitted with respect to one continuous wave light among the first optical coupler that separates the continuous wave light and the continuous wave light separated in the first optical coupler, the optical phase and the light Based on the second n-value data to be transmitted with respect to the other continuous wave light among the continuous wave light separated in the first optical modulator for changing the intensity and the first optical coupler, the optical phase and the light A second light modulation unit that changes the intensity, a phase shifter that performs a phase shift on the output light of either the first light modulation unit or the second light modulation unit, and the first A second optical coupler that multiplexes either the output light of the light modulation unit or the second light modulation unit and the output light of the phase shifter; and the first n-value data is zero level Sometimes the first data by which the output light of the first light modulator is quenched. A second bias for generating a second data bias voltage for quenching the output light of the second light modulation unit when the second n-value data is at a zero level. A bias power supply, an optical power monitor that outputs the optical power of the output light of the second optical coupler as an electrical signal, and a first dither signal having a predetermined first frequency is superimposed on the first data bias voltage. A first dither signal applying means; and a second dither signal having the same frequency as the first frequency in the second data bias voltage and having a phase orthogonal to the first dither signal. The second dither signal applying means for superimposing and the electric signal output from the optical power monitor are synchronously detected using a second frequency that is an integral multiple of the first frequency, and the integer is an even number of 2 or more. Yes, the synchronous detection result Phase amount correction means for correcting the phase shift amount in the phase shifter so as to be 0, and when the synchronous detection result is negative, the phase shift amount is increased, and when the synchronous detection result is positive, A positive slope mode for decreasing the phase shift amount, and a negative slope mode for increasing the phase shift amount when the synchronous detection result is positive, and decreasing the phase shift amount when the synchronous detection result is negative. An optical signal generation method in an optical transmitter comprising switching means for selecting any of the following:
A first step of determining the magnitude and sign of the control signal based on the synchronous detection result and the slope mode selected by the switching means;
And a second step of correcting the phase shift amount by feeding back the control signal to the phase shifter.

Images