JP5906166B2 - Driving method of light modulation element - Google Patents

Description

  The present invention relates to an optical modulator, and more particularly to a Mach-Zehnder optical modulator using light interference.

A modulator widely used at present is a Mach-Zehnder type optical modulator using lithium niobate (LiNbO ₃ ) which is a ferroelectric crystal. When an electric field is applied to an electro-optic crystal such as lithium niobate, its refractive index changes due to the Pockels effect, but the refractive index change is proportional to the applied electric field. In the Mach-Zehnder type optical modulator, a modulation operation such as ON / OFF of signal light is realized by combining a refractive index change caused by application of an electric field and a light interference effect.

  FIG. 1 is a diagram showing a configuration of a general Mach-Zehnder type light modulation element. The optical modulation element shown in FIG. 1 includes two arms 104a and 104b each including an input waveguide 101, an optical branching circuit 102, an optical combining circuit 103, an optical splitter circuit 102, and a waveguide connecting the optical combining circuit 103. , Phase modulators 105 a and 105 b installed on the arms 104 a and 104 b, and an output waveguide 106. In the light modulation element shown in FIG. 1, electrodes 107a and 107b are formed on phase modulators 105a and 105b. Electrodes 107a and 107b have electrode inlets 108a and 108b and electrode outlets 109a and 109b. ing. Usually, a microwave drive signal is applied to the inlets 108a and 108b, and is guided from the outlets 109a and 109b to a termination resistor connected to the outside of the device. The light is modulated as follows. That is, the signal light incident on the input waveguide 101 is branched into two by the optical branch circuit 102 to the arms 104a and 104b, and enters the phase modulators 105a and 105b, respectively. In the phase modulators 105a and 105b, the refractive index of the medium constituting the core or cladding of the waveguide is changed by the electric drive signals applied to the electrodes 107a and 107b, respectively, and as a result, the equivalent refractive index of the waveguide is changed. Thus, phase modulation of light propagating through the waveguide is performed. The phase-modulated signal light enters the optical converging circuit 103 in a state where the phase relationship between the arm 104a or 104b or the phase modulator 105a or 105b is modulated according to the path, and is guided to the output waveguide 106. . At this time, the intensity of the signal light incident on the output waveguide 106 changes according to the relative phase difference of the signal light that has passed through the respective arms due to light interference. This is the intensity modulation operation of the Mach-Zehnder modulator. In the light intensity modulation operation, it is necessary to pay attention to the phase change accompanying the intensity change, that is, the chirp.

  Here, a conventional method of driving the light modulation element will be described. The Mach-Zehnder optical modulator has a feature that chirp can be freely controlled by a driving method. This situation is summarized in FIG. FIG. 2 shows a diagram in which the horizontal axis represents time T, the vertical axis represents drive voltage V, modulated output light intensity P, and optical frequency shift Δf for each drive condition. Now, the upper arm 104a shown in FIG. 1 is called arm 1, and the lower arm 104b is called arm 2. For example, when only the phase modulator 105a installed in the arm 1 is driven by an electric signal (this electric signal is called "DATA1") (driving condition 1 in FIG. 2, one arm for driving only one arm) This is a negative chirp in which the optical frequency shifts in the negative direction at the rising portion of the output light intensity waveform and the optical frequency shifts in the positive direction at the falling portion of the output light intensity waveform. Further, when only the phase modulator 105b installed in the arm 2 is driven by an inverted signal of the drive signal used in the drive condition 1 (this electric signal is referred to as “DATA2”) (drive condition 2 in FIG. 2, this case) This is called single arm drive because it drives only one arm), and the optical frequency shifts in the positive direction at the rising edge of the output light intensity waveform, and the optical frequency shifts in the negative direction at the falling edge of the output light intensity waveform. It is possible to set the driving conditions such as positive chirp. What is important here is the driving condition of zero chirp in which the optical frequency shift does not occur at the rising and falling portions of the output light intensity waveform. This can be realized when the two phase modulators 105a and 105b are symmetrically driven by differential signals (that is, DATA1 and DATA2), that is, when push-pull driving is performed with equal voltage amplitude (driving condition 3 in FIG. 2). (For example, see Non-Patent Document 1 and Non-Patent Document 2).

  In Non-Patent Documents 1 and 2, in a symmetric Mach-Zehnder type modulator in which two arms have a symmetric structure as shown in FIG. 1, the same waveform and polarity, or the magnitude relationship between potentials at each level It is shown that when the push-pull drive is performed with two drive waveforms in which is inverted, a zero chirp operation is possible in the entire voltage range.

  Since it is a symmetrical push-pull that uses a differential signal, when the drive electrical signal applied to the phase modulator 105a installed in the arm 1 and the phase modulator 105b installed in the arm 2 is added, the amplitude is constant. Value. An example of the eye pattern of the output waveform of the modulator by symmetrical push-pull drive using a differential signal is shown in FIG. FIG. 3A shows a drive signal DATA1 applied to the phase modulator installed in the arm 1 and a drive signal DATA2 applied to the phase modulator installed in the arm 2 when the modulation speed is 10 Gbit / s. FIG. 3B shows the eye pattern of the sum of those, and FIG. 3B shows the eye pattern of the optical frequency and light intensity of the modulated light output of the Mach-Zehnder modulator. Since this is a symmetric push-pull for realizing zero chirp in which the optical frequency does not vary, the amplitudes of the DATA1 signal and DATA2 signal are equal, and the sum is a constant value including the bit transition region. Here, the bit transition region is a region where a level transition of the electric drive signal or the light intensity signal occurs between bits. Further, in order to bring the cross point to the position of 50% of the maximum value of the light intensity amplitude, the cross point of the electric waveform is also at the position of 50% of the amplitude in the eye pattern of the DATA1 signal and the DATA2 signal. Is required. Here, the cross point is a point where the rising edge and the falling edge intersect in the bit transition region of the electric drive signal waveform or the light intensity waveform. The driving method for generating a zero chirp signal in a Mach-Zehnder modulator that has been widely used in the past is the same amplitude and the sum of the amplitudes of the driving electric signals is constant by the differential signals DATA1 and DATA2 having a cross point of 50%. In other words, the push-pull drive is performed symmetrically. Under such driving conditions, as shown in FIG. 3B, the light intensity waveform has a cross point at a position of 50% of the maximum value of the light intensity amplitude, and a good modulation waveform with good symmetry is obtained. It can also be seen that the optical frequency shift is zero and has a constant value, that is, zero chirp operation is realized.

Here, supplementing the drive amplitude for the intensity-modulated output signal of the signal light to have a sufficient extinction ratio, the half-wave voltage _Vπ in the case of the one-arm drive of the drive conditions 1 and 2, the drive condition 3 In the case of symmetric push-pull driving, each differential signal may be driven with V _π / 2, which is half the half-wave voltage.

Further, while being a push-pull drive, the drive signal applied to the phase modulator installed in the arm 1 and the drive signal applied to the phase modulator installed in the arm 2 are asymmetric so that they are not equal in amplitude. When push-pull drive is performed, the chirp in the same direction as that when one-arm drive is performed on the phase modulator having a large amplitude of the applied signal among the phase modulators installed in the arms 1 and 2 respectively. Occurs. Therefore, for example, when the amplitude of the drive electric signal applied to the phase modulator 105a installed in the arm 1 is larger than the amplitude of the drive electric signal applied to the phase modulator 105b installed in the arm 2 Becomes negative chirp. However, the size (absolute value) of the chirp is smaller than that in the case of single arm drive. When the asymmetric push-pull drive is performed, the condition of the drive amplitude for the intensity-modulated output signal of the signal light to have a sufficient extinction ratio is the amplitude of the drive signal having a different amplitude for performing the asymmetric push-pull drive. The sum may be a half-wave voltage _Vπ .

  In the above example, the case where the arm 1 side is negatively chirped when one arm is driven has been described. However, whether the phase modulator connected to either arm 1 or arm 2 is driven to be negative chirp depends on the initial phase state, or an external drive electric signal (DATA1 above) And a bias voltage applied to the phase modulator of arm 1 or arm 2 in a superimposed manner and DATA2).

  The above is the operation of a general Mach-Zehnder type optical modulator. In the case of a modulator using a semiconductor material, the operation becomes complicated. (For example, refer nonpatent literature 3).

  Non-Patent Document 3 shows that in a Mach-Zehnder modulator using a semiconductor multiple quantum well structure, the α parameter is negative, that is, negative chirp even when symmetric push-pull driving is performed (non-chirp) (Non-Patent Document 3) Patent Document 3, FIG. 4).

  The reason for this will be described below. FIG. 4 shows an example of the relationship between applied voltage and refractive index change in a semiconductor. In addition to the Pockels effect in which the refractive index changes linearly with respect to the applied voltage (more precisely, the electric field), there is a component whose refractive index changes nonlinearly with respect to the applied voltage (more precisely, the electric field). This non-linear characteristic appears remarkably when, for example, the core layer of the optical waveguide is composed of multiple quantum well layers and the electro-optic effect due to the quantum confined Stark effect (QCSE) is used. The relative relationship between the Pockels effect and the QCSE is determined by the composition of the material, the film thickness, the magnitude of the electric field during use, and the like. When the refractive index change of the waveguide becomes nonlinear, the phase change of light propagating through the waveguide is also nonlinear. As described above, when a semiconductor is used for a part of the core or the clad of the optical waveguide, the phase change with respect to the applied voltage or the electric field becomes nonlinear, so that complete zero chirp operation is difficult. This situation is summarized in FIG. FIG. 5 shows a diagram of modulation characteristics when a semiconductor material is used, with the horizontal axis representing time T, the vertical axis representing drive voltage V, modulation output light intensity P, and optical frequency shift Δf for each driving condition. ing.

  When only the phase modulator 105a installed on the arm 104a in FIG. 1 (arm 1 in FIG. 5) is driven by the electric signal (DATA1) (driving condition 1 in FIG. 5), at the rising portion of the output light intensity waveform. The optical frequency shifts in the negative direction and the optical frequency shifts in the positive direction at the falling edge of the output light intensity waveform, and the phase modulator 105b installed in the arm 104b (arm 2 in FIG. 5). Is driven by the inverted signal (DATA2) of the drive signal used in drive condition 1 (drive condition 2 in FIG. 5), the optical frequency is shifted in the positive direction at the rising portion of the output light intensity waveform, and output is performed. The positive chirp that the optical frequency shifts in the negative direction at the falling edge of the light intensity waveform uses the ferroelectric crystal shown in FIG. It is similar to the modulator.

  However, in the case of a modulator using a semiconductor material, the phase modulators 105a and 105b are symmetrically driven by differential signals (that is, DATA1 and DATA2), that is, when push-pull drive is performed with an equal voltage amplitude (driving conditions in FIG. 5). 3), it shows a waveform like a differential waveform that rapidly shifts in the negative direction after the optical frequency shifts in the positive direction at the rising portion of the output light intensity waveform, and also at the falling portion of the output light intensity waveform. Similarly, it shows a waveform like a differential waveform similar to the rising portion, in which the optical frequency shifts in the positive direction and then rapidly shifts in the negative direction. As described above, the characteristic of the Mach-Zehnder type optical modulator using a semiconductor is that the same optical frequency change in the differential wave shape occurs at the rising and falling portions of the output light intensity waveform even during symmetric push-pull drive. is there. As shown in FIG. 5, this differential wave-shaped optical frequency shift is a frequency shift in the same direction as that of the negative chirp in the driving condition 1 in the portion where the output light intensity is sandwiched between the alternate long and short dash lines. In the case of a wave-shaped frequency shift, a negative chirp is substantially obtained.

  An example of an eye pattern of an output waveform of a Mach-Zehnder modulator using a semiconductor material by symmetrical push-pull drive using a differential signal is shown in FIG. 6A shows the drive signal DATA1 applied to the phase modulator installed in the arm 1, the drive signal DATA2 applied to the phase modulator installed in the arm 2, and the eye pattern of their sum. (B) shows the eye patterns of the optical frequency and light intensity of the modulated light output of the Mach-Zehnder modulator, respectively. The driving method for generating the zero chirp signal in the Mach-Zehnder modulator that has been widely used in the past, that is, the differential signals DATA1 and DATA2 having a cross point of 50%, the sum of the amplitudes of the driving electric signals is equal. This corresponds to a case where the driving method of performing the push-pull driving symmetrically so as to be constant is applied to the semiconductor Mach-Zehnder modulator.

  As shown in FIG. 6A, as in the case of FIG. 3, the amplitude of the DATA1 signal and DATA2 signal is the same as in the case of FIG. 3, and the sum is a constant value including the bit transition region. Also in the eye pattern of the DATA1 signal and the DATA2 signal, the cross point of the electric waveform is at a position of 50% of the amplitude. Under such driving conditions, the point where the rise and fall of the bit transition region of the light intensity waveform intersect, that is, the so-called cross point, is 50 as the maximum value of the light intensity amplitude as shown in FIG. % Position. This is also the same as the case of the modulator using the conventional dielectric material shown in FIG. However, when comparing the frequency shift, in the case of the modulator using the conventional dielectric material shown in FIG. 3, the frequency shift was zero and a constant value, whereas the semiconductor material of FIG. It can be seen that in the case of a modulator using, a differential frequency shift occurs in the bit transition portion. In other words, if no bit level transition occurs, no frequency shift occurs, so that it is flat at zero, whereas if a bit level transition occurs, the first half is a positive frequency shift and the second half is a negative frequency shift. Such a differential shape frequency shift occurs. In the case of using the dielectric material shown in FIG. 3B and the case of using the semiconductor material shown in FIG. There is a big difference in shifting. This indicates that it is impossible to determine the state of optical frequency shift, in other words, the state of frequency chirp, from the light intensity waveform.

  As described above, in the case of a Mach-Zehnder modulator using a semiconductor material, even when driven under the condition of zero chirp operation in a conventional dielectric modulator, the light intensity waveform can be obtained with a good symmetry, but the optical frequency A differential shape shift occurred and the zero chirp operation could not be performed.

  That is, in the case of a Mach-Zehnder modulator using a semiconductor material, even when symmetric push-pull driving is performed, a negative chirp is generated and a zero chirp operation cannot be realized.

G. L. Li et al., “Optical Intensity Modulators for Digital and Analog Applications”, Journal of Lightwave Technology, Vol.21, No.9, July 2003 pp.2010-2030 Osamu Mizuna, "Fiber-optics text for telecommunications chief engineer", ISBN978-4-89019-283-0, pp.184-191 John C. Cartledge, "Optimizing the Bias and Modulation Voltages of MQW Mach-Zehnder Modulators for 10 Gb / s Transmission on Nondispersion Shifted Fiber", Journal of Lightwave Technology, Vol.17, No.7, July 1999 pp.1142-1151 Oki Technical Review April 2002 / No.190 Vol.69 No.2 pp.68-71

  In a Mach-Zehnder modulator using a conventional semiconductor material, as described above, since the refractive index change or phase change during voltage application is non-linear, it becomes negative chirp even when symmetric push-pull drive is performed, and zero chirp operation Could not be realized. Further, even in this negative chirp state, the light intensity waveform has a cross point at a position of 50% of the maximum value of the light intensity amplitude, and it is difficult to judge the state of the chirp from the light intensity waveform, and means and method for adjusting the chirp There was almost no.

  In Non-Patent Document 3, it is proposed that the frequency chirp can be reduced by making the branching ratio of the waveguide branching circuit 102 constituting the Mach-Zehnder interferometer shown in FIG. However, it is difficult to shift the branching ratio from 1: 1 to a desired value, and it is difficult to say that it is a realistic method in consideration of the risk of wavelength dependency and the like. In Non-Patent Document 3, it has been proposed that the frequency chirp can be reduced by making the bias voltages of the phase modulators installed in the arms 1 and 2 asymmetrical. FIG. As can be seen from 4, sufficient effects cannot be obtained.

  The present invention solves the above-mentioned conventional problems and does not take unrealistic means such as changing the branching ratio from 1: 1. For example, the configuration used normally, for example, the light branching ratio is 1: 1. Provided is a driving method for realizing zero chirp in a Mach-Zehnder modulator in which the structures of two optical modulators are substantially the same, and an object thereof is to provide an optical transmission device using the driving method. .

To achieve the above object, according to the first aspect of the present invention, nonlinearity is applied to voltages applied to the first and second phase modulation regions respectively provided in the two arms constituting the light modulation element. A method for driving a Mach-Zehnder type optical modulation element having a favorable phase change characteristic, wherein a DATA1 signal is applied to a first phase modulation area and a DATA2 signal is applied to a second phase modulation area as a drive electric signal of the optical modulation element when, D and amplitude normalized respectively ATA1 signal at half-wave voltage of the first phase modulation region, rather equal and the amplitude normalized respectively DATA2 signal by half-wave voltage of the second phase modulation region, The DATA1 signal is logically inverted to become a DATA2 signal, and the sum of the applied voltages of the driving electric signals of the DATA1 signal and the DATA2 signal takes a constant value except for the bit transition region, The bit transition region is set to be larger than the predetermined value . Thus, even in a Mach-Zehnder modulator using a semiconductor material having a non-linear effect, a zero chirp operation is possible.

According to the second aspect of the invention, the bit width of the mark level where the absolute value of the applied voltage of each of the DATA1 signal and the DATA2 signal is large is wider than the bit width of the space level where the absolute value of the applied voltage is small. by setting, the sum of the applied voltage of the driving electrical signals DATA1 signal and DATA2 signal takes a constant value except bit transition region, and a bit transition region is set to be larger than a certain value. Thus, even in a Mach-Zehnder modulator using a semiconductor material having a non-linear effect, a zero chirp operation is possible.

In the invention according to claim 3, in the eye patterns of the applied voltages of the DATA1 signal and the DATA2 signal, the applied voltage is applied in the transition region between the mark level where the absolute value of the applied voltage is large and the space level where the absolute value of the applied voltage is small. rising and falling of is set to shift to the large direction of the absolute value than the center value of the cross point are all applied voltage crossing. Thus, even in a Mach-Zehnder modulator using a semiconductor material having a non-linear effect, a zero chirp operation is possible.

In the invention according to claim 4, when the maximum value in the bit transition region of the sum of the applied voltages of the DATA1 signal and the DATA2 signal is Emax and the constant value in the region excluding the bit transition region is Econst, Re = Emax / Econst is set in the range of 1 <Re <1.67. Thus, even in a Mach-Zehnder modulator using a semiconductor material having a non-linear effect, a zero chirp operation is possible.

According to the fifth aspect of the present invention, when the bit width of the mark level with a large applied voltage of the DATA1 signal and the DATA2 signal is Tmk and the bit width of the space level with a small applied voltage is Tsp, the ratio is Rt = Tmk / Tsp. Is set in the range of 1 <Rt <1.7. Thus, even in a Mach-Zehnder modulator using a semiconductor material having a non-linear effect, a zero chirp operation is possible.

According to the sixth aspect of the present invention, when the position of the cross point of the applied voltage of the DATA1 signal and the DATA2 signal is Rx (%), Rx is set in the range of 50 <Rx <84. Thus, even in a Mach-Zehnder modulator using a semiconductor material having a non-linear effect, a zero chirp operation is possible.

  According to the present invention, even in a Mach-Zehnder modulator using a semiconductor material having a non-linear effect typified by the QCSE effect, zero chirp operation is possible, and a small and high-performance optical transmitter can be realized. . In addition, since the QCSE effect can be used positively, it is possible to further reduce the drive voltage and power consumption of the optical modulator and the optical transmitter using the same.

It is a structural diagram showing a configuration example of a conventional Mach-Zehnder modulator. It is a conceptual diagram in which the horizontal axis represents time T, the vertical axis represents drive voltage V, modulation output light intensity P, or optical frequency shift Δf with respect to each drive condition of a conventional dielectric modulator. It is a figure of the result of the simulation which shows an example of the eye pattern of the drive waveform and modulation | alteration output waveform of the conventional dielectric modulator, (a) is DATA1 signal, DATA2 signal, and these sum, (b) is modulated light. It is a figure of the result of the simulation which shows the eye pattern of output optical frequency and light intensity. It is a graph of the measurement data which shows the relationship between the applied voltage of a semiconductor waveguide, and a refractive index change. It is a conceptual diagram in which the horizontal axis represents time T, the vertical axis represents drive voltage V, modulation output light intensity P, and optical frequency shift Δf for each drive condition of a conventional semiconductor modulator. It is a figure of the result of the simulation which shows an example of the drive waveform of the conventional semiconductor modulator, and an eye pattern of a modulation output waveform, (a) is DATA1 signal, DATA2 signal, and these sum, (b) is modulated light output. It is a figure of the result of the simulation which shows the eye pattern of optical frequency and light intensity of. FIG. 5 is a conceptual diagram in which the horizontal axis represents time T, the vertical axis represents drive voltage V, modulated output light intensity P, and optical frequency shift Δf for each drive condition of the present invention. It is a figure of the result of the simulation which shows the eye pattern of the drive waveform of this invention, and a modulation | alteration output waveform, (a) is DATA1 signal, DATA2 signal, and these sum, (b) is the optical frequency and light of modulated light output. It is a figure of the result of the simulation which shows a strong eye pattern. It is an enlarged view of an eye pattern of a drive waveform of the present invention, (a) is a diagram explaining the definition of the cross point of the mark level and the space level, (b). It is a figure explaining the definition of the eye pattern of the drive waveform of this invention, (a) is a figure explaining the definition of the eye pattern when a drive voltage is positive, (b) when a drive voltage is negative. FIG. 4 of Non-Patent Document 4. FIG. 10 is a structural diagram showing a configuration example when a new electrode is provided on an arm of a conventional Mach-Zehnder modulator. An example of the relationship between the maximum value Emax in the summed bit transition region in FIG. 8A, the ratio Re of the constant value Econst in the region excluding the bit transition region, and the peak value of the optical frequency shift in FIG. It is a figure of the result of the shown simulation. FIG. 9 is a simulation result showing an example of the relationship between the mark level bit width Tmk and the space level bit width Tsp ratio Rt in FIG. 9A and the peak value of the optical frequency shift in FIG. is there. It is the figure of the result of the simulation which showed an example of the relationship between the position of the cross point in FIG.9 (b), and the peak value of the optical frequency shift in FIG.6 (b).

[Example]
FIG. 7 is a diagram for explaining an embodiment of the present invention. With respect to the modulation characteristics of a Mach-Zehnder type optical modulation element using a semiconductor material having the configuration as shown in FIG. The figure shows one of the drive voltage V, the modulated output light intensity P, and the optical frequency shift Δf. Conventional drive in that push-pull drive is performed on the phase modulators 105a and 105b respectively provided in the two arms constituting the Mach-Zehnder modulator, the DATA1 signal, and the DATA2 signal logically inverted with respect to the DATA1 signal. Same as the law.

  The features of the driving method of the present application are as follows. In the DATA2 signal, a signal logically inverted with respect to the DATA1 signal is changed to the DATA2 signal, and the sum of the applied voltages of the DATA1 signal and the DATA2 signal is constant except for the bit transition region and the bit transition The area is set to be larger than the predetermined value. Outside the bit transition region, the amplitudes of the drive electric signals of the DATA1 signal and the DATA2 signal are equal, and symmetric push-pull drive is performed. In the conventional symmetrical push-pull drive, the sum of the applied voltages is constant including the bit transition region, whereas in the driving method of this application, the symmetrical push-pull drive except for the bit transition region and the symmetry in the bit transition region. The point of breaking down is very different.

  This is realized by setting the bit width of the mark level where the absolute value of the applied voltage of the drive electric signal of the DATA1 signal and the DATA2 signal is larger than the bit width of the space level where the absolute value of the applied voltage is small. It is possible. That is, as shown in FIG. 7, the phase modulator 105a installed in the arm 1 is driven by the DATA1 signal, the phase modulator 105b installed in the arm 2 is driven by the DATA2 signal, and the DATA1 signal and DATA2 are driven. Both of the signals are set so that the width of one bit of the mark level where the absolute value of the voltage of the drive signal is large is wider by the same ratio than the one bit width of the portion where the output light intensity is sandwiched between the dashed lines. This can be realized. At this time, the bit width of the space level where the absolute value of the applied voltage is small is automatically narrower than the bit width of the mark level.

  When such driving conditions are used, the optical modulation output is greatly reduced so that the optical frequency shift becomes almost zero as shown in FIG. 7 while maintaining the cross-point of the intensity waveform at 50%. The frequency chirp can be greatly reduced. Here, the cross point is defined as a point where the rising edge and the falling edge intersect in the transition region of the bit of the electric drive signal waveform or the light intensity waveform.

  An example of an eye pattern of an output waveform of a Mach-Zehnder modulator using a semiconductor material when the driving method of the present invention is used is shown in FIG. FIG. 8A shows a drive signal DATA1 applied to the phase modulator installed in the arm 1, a drive signal DATA2 applied to the phase modulator installed in the arm 2, and an eye pattern of their sum. 8 (b) shows an eye pattern of the optical frequency and light intensity of the modulated light output of the Mach-Zehnder modulator, respectively.

  In the driving conditions shown in FIGS. 3 and 6, the DATA1 signal that is a driving signal applied to the phase modulator installed in the arm 1 and the driving signal applied to the phase modulator installed in the arm 2 are used. The sum of the drive amplitudes of the DATA2 signal was a constant value including the bit level transition region. However, in the driving method of the present application shown in FIG. 8, when the bit level transition does not occur, the sum of the drive amplitudes is flat and constant, while when the bit level transition occurs, the drive amplitude is increased in the transition region. The sum is larger than a certain portion excluding the bit transition region. Looking at the eye patterns of the DATA1 signal and the DATA2 signal at this time, as shown in FIG. 8, the amplitude of the drive electrical signal is the same for the DATA1 signal and the DATA2 signal. That is, the DATA1 signal and the DATA2 drive electric signal having the same amplitude are applied to the DATA1 signal and the DATA2 signal so that the sum of the applied voltages is constant except for the bit transition region and larger than the constant value in the bit transition region. It is the first feature of the present invention that the drive electric signal waveform of the DATA2 signal is set. Thus, as shown in FIG. 8B, it is possible to realize a zero chirp operation in which the cross point of the light intensity waveform is 50% and the frequency shift hardly occurs including the bit transition region.

  Further, as shown in FIG. 9A, both the DATA1 signal and the DATA2 signal are wider by the same ratio than the 1-bit width of the “mark level” than the 1-bit width of the “space level”. . Here, as shown in FIG. 9A, the “mark level” refers to a level where the voltage of the drive signal is larger than the half value of the drive signal, and the “space level” refers to the voltage of the drive signal The level is smaller than the half value of the driving signal. Here, the bit width will be described in detail. FIG. 9 shows an enlarged view of the eye pattern of the drive electrical waveform. As shown in FIG. 9A, it is assumed that the “space level” potential is 0, the “mark level” potential is 1, and the maximum voltage is 1. In the transition region of the level of the drive electric signal, the level of the drive signal at the first fall after the rise from the point that the drive signal level at the rise is 50% of the maximum amplitude, that is, 0.5 after the standardization. The width up to a point where 50% of the maximum amplitude, that is, 0.5 after normalization, is defined as the width of the “mark level”. Conversely, from the point that the drive signal level at the fall is 50% of the maximum amplitude, that is, 0.5 after the standardization, the drive signal level at the rise that first appears after the fall is 50% of the maximum amplitude, that is, the standard. The width up to the point of 0.5 after conversion is defined as the “space level” width. At this time, the width of the “mark level” is wider than the width of the “space level”. The second feature of the present invention is that the “mark level” width of the DATA1 signal and the DATA2 signal is set to be wider than the width of the “space level”, and is matched in both the DATA1 signal and the DATA2 signal. is there. By setting the drive electric signals of the DATA1 signal and the DATA2 signal having the same amplitude in this way, the cross point of the light intensity waveform is 50% and the frequency shift is a bit transition region as shown in FIG. It is possible to realize a zero chirp operation including almost no.

  Another way of looking at the drive electrical waveform of FIG. 9 is also possible. This will be described with reference to FIG. FIG. 9B shows the driving waveform shown in FIG. 9A again, but this time, the cross point where the rise and fall intersect in the transition region of “mark level” and “space level”. Pay attention to.

In the driving waveform shown in FIG. 9, the cross point is larger than 50% of the maximum value of the amplitude of the drive electric signal, and the cross point is shifted in a direction in which the absolute value is larger than the center value of the modulation applied voltage. It is set to be. Further, in the driving method of the present application, as shown in FIG. 8A, in both the DATA1 signal and the DATA2 signal, the position of the cross point is shifted in a direction in which the absolute value is larger than the center value of the modulation applied voltage. And the shift amount is set so as to match between the DATA1 signal and the DATA2 signal. Here, the meaning of the direction in which the absolute value is large will be described with reference to FIG. FIG. 10 shows the waveform (eye pattern) of an electric signal applied as a drive signal. FIG. 10A shows a case where the phase change shown in FIG. 4 occurs when the drive voltage to the element is positive, that is, when a positive voltage is applied. FIG. 10B shows a case where the phase change as shown in FIG. 4 occurs when the drive voltage is negative, that is, when a negative voltage is applied. Usually, in the case of an optical modulator using a semiconductor material, a method is employed in which a high-frequency data signal corresponding to a data bit rate is superimposed on a DC bias voltage to drive an element. Of these, the portion of the drive voltage based on the high-frequency data signal excluding the bias voltage component corresponds to the drive electrical signal based on the DATA1 signal and DATA2 signal described so far in the present invention. Considering the polarity of the drive voltage, when the point where the absolute value of the applied voltage of the drive signal including the bias voltage is the smallest is 0 and the point where the absolute value of the voltage is the largest is V _m , the high frequency DATA1 The amplitude of the drive signal by the signal and the DATA2 signal is | V _m |. When expressed in this way, the meaning of the direction with a larger absolute value in the present application means that 0.5 V _m <V <in the case of the driving condition in the positive direction shown in FIG. 10A (ie, V _m > 0). The voltage V of V _m indicates the voltage V of V _m <V <0.5 V _{m in} the negative driving condition shown in FIG. 10B (ie, V _m <0). When the cross point is within this range, the position of the cross point measured in the direction from the 0 point to the V _m point is shifted in a direction in which the absolute value is larger than the center value of the maximum amplitude modulation applied voltage. FIG. 10 shows an example of a waveform when the cross point is 60%.

  Conventionally, as a method for correcting nonlinear absorption characteristics of a semiconductor, a method of shifting a cross point in a drive waveform of an electroabsorption modulator (so-called EA modulator) has been adopted. (Non-Patent Document 4) According to this, “the output waveform of this driver is adjusted so that the cross point is about 70% so that the cross point of the optical output signal is 50%” is described. ing. In order to compare this with this application, FIG. 4 of the nonpatent literature 4 is made into FIG. 11 of this application, and it demonstrates using FIG. In the EA modulator, if a driving waveform having a cross point at a position of 50% of the maximum amplitude is used, the cross point of the modulated output light intensity waveform is shifted in the space level direction, that is, in a direction smaller than 50%. For the purpose of adjusting the cross point of the optical output signal to 50%, the cross point of the drive waveform is shifted. According to Non-Patent Document 4, it is “about 70%”. However, since the EA modulator applies a negative voltage, the definition used in this application, that is, the absolute value counted from the smaller absolute value side of the amplitude. When expressed based on the definition of what percentage of the maximum value is, it corresponds to about 30% as shown in FIG. Comparing this application with Non-Patent Document 4, the purpose of Non-Patent Document 4 is to adjust the cross point of the modulation output light intensity waveform of the EA modulator to 50% as shown in FIG. In the case of a Mach-Zehnder modulator in which the cross point of the output light intensity waveform is 50% without adjusting the cross point of the electrical waveform, it is not necessary to adjust the cross point of the drive electrical waveform. Furthermore, in the case of Mach-Zehnder modulation, if the cross point of the drive electrical waveform is set to a value other than 50% in the symmetric push-pull drive, the cross point of the modulated output light intensity waveform also deviates from 50%. Therefore, the concept of symmetrical push-pull drive, which has been considered essential in the past, and the concept of adjusting the cross point of the drive electrical waveform are incompatible with each other. As for the direction of the desired driving electric signal waveform cross point shift, in the present application, the cross point is set to shift in a direction having a larger absolute value than the center value of the modulation applied voltage, that is, a direction larger than 50%. On the other hand, in Non-Patent Document 4, since the cross point is set to shift in a direction having a smaller absolute value than the center value of the modulation applied voltage, that is, a direction smaller than 50%, the directions are also opposite. Further, in Non-Patent Document 4, the purpose is to adjust the cross point of the modulation output light intensity waveform of the EA modulator to 50%, and at this time there is no means for controlling the frequency chirp to a desired state. In Patent Document 4, it is clear that no consideration is given to frequency chirp. Based on the above facts, regarding a Mach-Zehnder modulator using a semiconductor material having a non-linear phase change disclosed in the present application, a driving method for controlling frequency chirp with a simple configuration and realizing zero chirp operation is provided. It is clear that Non-Patent Document 4 cannot give any suggestion regarding technology.

  As described above, in the driving method of the Mach-Zehnder modulator, the amplitude that was first derived by abandoning the concept of completely symmetric push-pull driving, which is the conventional common sense, and introducing the concept of breaking symmetry in the bit transition region. Are set such that the positions of the cross points of the drive electric signals of the DATA1 signal and the DATA2 signal having the same value are shifted in the direction in which the absolute value is larger than the center value of the modulation applied voltage, and the shift amount is the same as the DATA1 signal and the DATA2 signal The third feature of the driving method of the Mach-Zehnder optical modulator disclosed for the first time in the present invention is that the values are set to match. By setting the drive electric signals of the DATA1 signal and the DATA2 signal in this way, the cross point of the light intensity waveform is 50% and the frequency shift includes the bit transition region as shown in FIG. 8B. It is possible to realize a zero chirp operation that hardly occurs.

  As another example of the method for shifting the cross point, the cross point is always desired by using a function of a known modulator driver that adjusts the cross point to the center when a distorted waveform is input. There is also a method of setting so as to be shifted to the position of.

In the above description, the case of a Mach-Zehnder modulator having a symmetric structure has been described. The half-wave voltages Vπ of the devices may be different from each other. In that case, in the above description, “the amplitude of the DATA1 signal and the DATA2 signal is equal” is replaced with “the amplitude normalized by the half-wave voltage Vπ of the phase modulator to which the DATA1 signal and the DATA2 signal are applied, respectively. The above description can be applied as it is. In other words, the amplitude of the DATA1 signal V _data, the amplitude of the DATA2 signal V _DataBar, DATA1 signal half-wave voltage amplitude of the phase modulator half-wave voltage amplitude is applied Vpai_ _data, DATA2 signal of the applied phase modulator when is the _{Vπ_} databar, is replaced with V _data / Vπ_ _data and V _databar / _{Vπ_ databar} are equal. Furthermore, in the case of a Mach-Zehnder modulator using a material having nonlinear phase modulation characteristics such as a semiconductor material, different bias voltages are applied to the phase modulators installed in the respective arms in order to set an appropriate operating point. When used in an applied state, the half-wave voltages Vπ of the two phase modulators may be different from each other even in a Mach-Zehnder modulator having a symmetric structure. Also in this case, as in the case of the asymmetric Mach-Zehnder modulator, in the above description, “the amplitude of the DATA1 signal and the DATA2 signal is equal” is replaced with “half of the phase modulator to which the DATA1 signal and the DATA2 signal are applied, respectively. It is possible to apply the above description as it is by reading as “the amplitudes normalized by the wavelength voltage Vπ are equal”.

  On the other hand, in the case of a Mach-Zehnder modulator using a material having non-linear phase modulation characteristics such as a semiconductor material, it does not occur that the half-wave voltages Vπ of the two phase modulators are different from each other when the bias point is set. Therefore, a new electrode may be provided on the arm. FIG. 12 shows the configuration. The Mach-Zehnder modulator shown in FIG. 12 includes two arms 204a and 204b each including an input waveguide 201, an optical branching circuit 202, an optical combining circuit 203, an optical splitter circuit 202, and a waveguide connecting the optical combining circuit 203. , Phase modulators 205 a and 205 b installed on the arms 204 a and 204 b, and an output waveguide 206. It also has electrodes 207a and 207b for phase modulators 205a and 205b, electrode inlets 208a and 208b, electrode outlets 209a and 209b, and typically a microwave drive signal is applied to inlets 208a and 208b, and outlets 209a and 208b 209b leads to a termination resistor connected to the outside of the element. Phase adjustment regions 210a and 210b installed on two arms 204a and 204b, respectively, and phase adjustment electrodes 211a and 211b are further provided. The operation principle of the element of this configuration is to understand the operation of the modulation element of the configuration shown in FIG. 12 as it is by replacing 101 to 109 in the description of the operation principle of the modulation element having the structure shown in FIG. 1 with 201 to 209. Is possible. In the configuration shown in FIG. 12, since the phase adjustment regions 210a and 210b are newly provided, the operating point of the Mach-Zehnder modulator is set to the phase adjustment electrode 211a while the bias voltages of the phase modulators 205a and 205b are set to the same value. And 211b can be adjusted by a voltage applied to or both of them and a current to be injected. Therefore, if the Mach-Zehnder modulator having this configuration is used, the effects of the present invention can be realized ideally in a state where the half-wave voltages Vπ of the two phase modulators 205a and 205b do not differ from each other.

  Finally, the most desirable conditions where the effect of the present invention can be expected will be described.

  FIG. 13 shows that when the maximum value in the bit transition region of the sum of the voltages applied to the drive electric signals of the DATA1 signal and the DATA2 signal in FIG. 8A is Emax, and the constant value in the region excluding the bit transition region is Econst. FIG. 6B shows an example of the relationship between Re = Emax / Econst, which is the ratio of the above, and the peak value of the optical frequency shift in FIG. 6B, where Re = 1.0 is the conventional completely symmetric push-pull drive. It corresponds. In the vertical axis of FIG. 13, the sign of the optical frequency shift corresponds to negative chirp and positive to positive chirp, and the smaller the absolute value, the smaller the chirp. Therefore, when the driving method of the present invention is used for zero chirping, the range in which the chirp is reduced can be regarded as a range in which the absolute value of the frequency shift in FIG. 6 is smaller than that when Re = 1.0. It can be seen that 0.0 <Re <1.67. That is, if a driving condition such that Re is in the above range is used, chirp can be reduced, and the effects of the present invention can be obtained.

  FIG. 14 shows the ratio of the bit width of the mark level with the large applied voltage of the drive electric signal of the DATA1 signal and the DATA2 signal in FIG. 9A as Tmk and the bit width of the space level with the small applied voltage as Tsp. FIG. 7 is a diagram showing an example of the relationship between a certain Rt = Tmk / Tsp and the peak value of the optical frequency shift in FIG. 6B, where Rt = 1.0 corresponds to the conventional completely symmetric push-pull drive. Yes. The sign of the optical frequency shift on the vertical axis in FIG. 14 corresponds to negative chirp and positive to positive chirp. The smaller the absolute value, the smaller the chirp. Therefore, when the driving method of the present invention is used for zero chirping, the range in which the chirp is reduced can be said to be a range in which the absolute value of the frequency shift is smaller than when Rt = 1.0 in FIG. It can be seen that 0.0 <Rt <1.7. That is, if a driving condition in which Rt is in the above range is used, the chirp can be reduced, and the effects of the present invention can be obtained.

  FIG. 15 is a diagram showing an example of the relationship between the position of the cross point of the driving electric signal of the DATA1 signal and the DATA2 signal in FIG. 9B and the peak value of the optical frequency shift in FIG. = 50% corresponds to the conventional fully symmetrical push-pull drive. As for the sign of the optical frequency shift on the vertical axis in FIG. 15, minus corresponds to negative chirp and plus corresponds to positive chirp. The smaller the absolute value, the smaller the chirp. Therefore, when the driving method of the present invention is used for zero chirping, the range in which the chirp is reduced can be regarded as a range that is smaller than when the absolute value of the frequency shift is Rx = 50% in FIG. It can be seen that Rx <84. That is, if a driving condition is used such that Rx is in the above range, chirp can be reduced, and the effects of the present invention can be obtained.

  FIGS. 13, 14, and 15 show examples in which QCSE, which is a non-linear effect in FIG. 4, is dominant. Even when non-linear contribution is small, optimum values of Re, Rt, and Rx are shown. Falls within the above range. The ratio of the contribution of the nonlinear effect varies depending on the arm of the Mach-Zehnder modulator or the structure of the waveguide forming the phase modulator installed in the arm. However, if Re, Rt, or Rx is within the above range, Even when a modulator having a structure is used, the effect of the present invention can be obtained.

As described above, the driving method of the Mach-Zehnder modulation element has been described by taking the intensity modulation by the NRZ signal as an example, but the driving method of the present invention is also effective for other modulation methods. For example, the optical duobinary method can be handled by setting the amplitude of the drive electric signal to an amplitude necessary for optical duobinary modulation. Specifically, in the case of NRZ intensity modulation, push-pull drive is performed at half of the half-wave voltage, whereas push-pull drive is performed at half-wave voltage that is twice that of NRZ intensity modulation. If set, it is possible to support optical duobinary modulation. The same applies not only to intensity modulation but also to phase modulation operations represented by DPSK modulation.

  In the above embodiment, a normal Mach-Zehnder modulator having two arms has been described. However, the two arms of a so-called parent Mach-Zehnder interferometer are called sub-Mach-Zehnder modulator or child Mach-Zehnder modulator, respectively. For the so-called quadrature modulator or IQ modulator in which two Mach-Zehnder modulators are arranged as described in the example, the driving method according to the above embodiment is applied to each sub-Mach-Zehnder modulator or child Mach-Zehnder modulator. Thus, it goes without saying that the effect of the nonlinear refractive index change, which is the effect of the present invention, can be reduced.

  Furthermore, application to multilevel modulation such as QPSK modulation and QAM is also possible.

101, 201 Input waveguide 102, 202 Optical branch circuit 103, 203 Optical converging circuit 104a, 104b, 204a, 204b Arm 105a, 105b, 205a, 205b Phase modulator 106, 206 Output waveguide 107a, 107b, 207a, 207b Phase Modulators 108a, 108b, 208a, 208b Electrode inlets 109a, 109b, 209a, 209b Electrode outlets 210a, 210b Phase adjustment regions 211a, 211b Electrodes

Claims (6)

A method for driving a Mach-Zehnder type optical modulation element having a nonlinear phase change characteristic with respect to voltages applied to first and second phase modulation regions respectively provided in two arms constituting the optical modulation element,
When a DATA1 signal is applied to the first phase modulation region and a DATA2 signal is applied to the second phase modulation region as drive electric signals of the light modulation element ,
The amplitude normalized respectively DATA1 signal at half-wave voltage of the first phase modulation region, amplitude and is rather equal to the DATA2 signal respectively normalized by the half-wave voltage of the second phase modulation region,
The DATA1 signal is logically inverted to become the DATA2 signal, and the sum of the applied voltages of the driving electric signals of the DATA1 signal and the DATA2 signal takes a constant value except for the bit transition area, and the constant value in the bit transition area A method for driving a light modulation element, characterized in that the light modulation element is set to be larger . By setting the bit width of the mark level where the absolute value of the applied voltage of each of the DATA1 signal and the DATA2 signal is larger than the bit width of the space level where the absolute value of the applied voltage is small , the DATA1 signal And the sum of the applied voltages of the drive electric signals of the DATA2 signal is set to take a constant value except for the bit transition region and to be larger than the constant value in the bit transition region. Item 8. A method for driving a light modulation element according to Item 1. In the eye pattern of each applied voltage of the DATA1 signal and the DATA2 signal, the rising edge of the applied voltage in a transition region between a mark level where the absolute value of the applied voltage is large and a space level where the absolute value of the applied voltage is small 3. The method of driving an optical modulation element according to claim 1, wherein all of the cross points at which falling crosses are set to shift in a direction in which an absolute value is larger than a center value of the applied voltage. . When the maximum value in the bit transition region of the sum of the applied voltages of the DATA1 signal and the DATA2 signal is Emax and the constant value in the region excluding the bit transition region is Econst, the ratio Re = Emax / Econst is The method of driving a light modulation element according to claim 1, 2 or 3 , wherein 1 <Re <1.67. When the bit width of the mark level with the large applied voltage of the DATA1 signal and the DATA2 signal is Tmk and the bit width of the space level with the small applied voltage is Tsp, the ratio Rt = Tmk / Tsp is 1 <Rt < The method of driving a light modulation element according to claim 2 , which is in a range of 1.7. 4. The light modulation element according to claim 3 , wherein Rx is in a range of 50 <Rx <84, where Rx (%) is a position of a cross point of the applied voltage of the DATA1 signal and the DATA2 signal. Driving method.