JP6385848B2 - Light modulator - Google Patents

Description

  The present invention relates to an optical modulator applicable to an optical communication system.

  One of the means for realizing high-speed communication in an optical fiber communication system is optical time division multiplexing (OTDM).

A typical configuration of a conventional OTDM optical transmitter is that an optical pulse train is branched into N channels, the branched optical pulse trains are modulated with independent data by respective optical modulators, and then time delayed between channels by an optical delay line. An OTDM signal is obtained by giving a difference and finally combining (see Non-Patent Document 1 and Non-Patent Document 2).
If the drive symbol rate of each optical modulator is B (sps: symbol / second), the repetition frequency of the input optical pulse train is B, and the delay time provided by the optical delay line is 1 / (NB) with respect to the adjacent channel. The symbol rate of the final output optical signal is NB.

  As an OTDM optical transmitter different from the above, an input CW (Continuous Wave) light is periodically divided into two paths using a 1 × 2 optical switch circuit driven by a clock signal, and each of the distributed lights is distributed. A configuration has also been proposed in which an OTDM signal is obtained by multiplexing after modulation with independent data in a path (see Patent Document 1, Patent Document 2, and Non-Patent Document 3).

Here, an example of a conventional OTDM optical modulator in which the number of channels N = 2 is described with reference to FIG.
As shown in the figure, the light source 1 outputs continuous light in a single longitudinal mode. The first light modulation means (1 × 2 optical switch circuit) 2 modulates the continuous light output from the light source 1 with the clock signal having the clock frequency B. Thereby, a return-to-zero (RZ) optical pulse is alternately transmitted from each output port of the first optical modulation means 2. At this time, the repetition frequency of the RZ optical pulse train output to each port is B, and the peak timing of the optical pulses is shifted by 1 / (2B) between the output ports. That is, a pair of optical pulses having opposite polarities are output from each output port of the first light modulator 2.

Of the pair of light pulses output from the first light modulation means 2, one is input to the second light modulation means 3 and the other is input to the third light modulation means 4. A pair of data signals that are 180 degrees out of phase and synchronized with the clock signal input to the first optical modulation unit 2 are input to the optical modulation units 3 and 4.
As a result, the second optical modulation means 3 optically modulates one optical pulse input from the first optical modulation means 2 with a data signal of a symbol rate B that matches the peak timing of the one optical pulse. Then, a modulated light pulse is output. The third optical modulation unit 4 optically modulates the other optical pulse input from the first optical modulation unit 2 with a data signal of a symbol rate B that matches the peak timing of the other optical pulse. The modulated light pulse is output.

  The modulated optical pulse output from the second optical modulation means 3 and the modulated optical pulse output from the third optical modulation means 4 are combined by the optical coupling means 5 to obtain the final optical output (output optical signal). It becomes. The symbol rate of the final optical output (output optical signal) output from the optical coupling unit 5 is 2B, and the spectral width is 4B.

Japanese Patent No. 2823872 JP-A-10-79705

S. Kawanishi, “Ultrahigh-Speed Optical Time-Division-Multiplexed Transmission Technology Based on Optical Signal Processing,” IEEE J. Quantum Electron., Vol. 34, no. 11, pp. 2064-2079 (1998). HG. Weber, R. Ludwig., S. Ferber, C. Schmidt-Langhorst, M. Kroh, V. Marembert, C. Boerner, and C. Schubert, “Ultrahigh-Speed OTDM-Transmission Technology,” J. Lightw. Technol., Vol. 24, no. 12, pp. 4616-4627 (2006). G. Ishikawa, H. Ooi, Y. Akiyama, S. Taniguchi, and H. Nishimoto, “80-Gb / s (2 × 40-Gb / s) Transmission Experiments Over 667-km Dispersion-Shifted Fiber Using Ti: LiNb03 OTDM Modulator And Demultiplexer, ”Proc. ECOC 1996, paper ThC.3.3 (1996). M. Nakazawa, T. Hirooka, P. Ruan, and P. Guan, “Ultrahigh-speed“ orthogonal ”TDM transmission with an optical Nyquist pulse train,” Opt. Express, vol. 20, no. 2, pp. 1129- 1140 (2012).

In a typical conventional configuration as shown in Non-Patent Document 1 and Non-Patent Document 2 above, a pulse of a mode-locked laser or the like is used as a method for generating an optical pulse train so that optical pulses do not overlap in time between channels. Use a light source.
As such a pulse light source, a light source having a pulse width sufficiently narrower than the symbol time interval 1 / (NB) of the final output optical signal is used, so that the spectrum of the output optical signal spreads and the spectrum utilization efficiency is low. There was a problem of becoming. The specific spectral width varies depending on the type of light source and the use conditions, but at least the non-return-to-zero (NRZ) signal width of the symbol rate NB 2 NB (defined as the width between null points closest to the carrier frequency) Wider than.
Furthermore, since the oscillation wavelength and repetition frequency of the pulse light source are usually fixed for each light source device, there has been a problem that the transmission wavelength and symbol rate cannot be selected flexibly.

As a method for improving the spectrum utilization efficiency, a method has been proposed in which orthogonal Nyquist optical pulses are used to reduce the spectral width to the same level as the symbol rate NB of the final output optical signal without causing intersymbol interference ( Non-patent document 4).
However, this method requires a device for shaping an optical pulse into a Nyquist pulse shape (Pulse Shaper in FIG. 5 of Non-Patent Document 4) separately from the pulse light source, and the device configuration of the entire transmitter is complicated. It was.
Also, since the pulse light source is used, there is a problem that the wavelength and symbol rate cannot be selected flexibly.

In the conventional configurations shown in Patent Document 1, Patent Document 2 and Non-Patent Document 3 described above, since the RZ optical pulse having the repetition frequency B is used, the spectrum width of the final output optical signal (null point closest to the carrier frequency) Defined as the width in between) is about 4B, which is twice the final output symbol rate 2B.
For this reason, it is possible to achieve better spectrum utilization efficiency than the conventional configuration of Non-Patent Document 1, and because the CW light source is used, the wavelength and symbol rate can be selected flexibly, and an external pulse shaping device or the like is not required. Therefore, it can be said that the apparatus configuration is simple.
However, the spectrum width of the final output optical signal is about 4B. Thus, there is room for further improvement in that the spectrum width of the output optical signal is wide.

  The present invention has been made in view of such problems, and an object thereof is an OTDM system in which the spectrum width of an output optical signal is narrow, the wavelength and symbol rate can be selected flexibly, and the configuration is simple. An optical modulator is provided.

The present invention that achieves such an object ,
One optical port for input,
A 1-input 2-output optical coupler optically connected to the one input optical port;
Two pulse generators optically connected to two outputs of the one-input two-output optical coupler;
A two-input two-output optical coupler optically connected to respective outputs of the two pulse generators;
Two data modulators optically connected to the two outputs of the two-input two-output optical coupler;
An optical multiplexing unit optically connected to the respective outputs of the two data modulation units,
The two data modulators are each driven by a data signal with a symbol rate B, and the two pulse generators are respectively driven by signals having a periodic waveform with a frequency B / 2 having a phase different from each other by π / 2,
When the optical path length from the input optical port to one of the outputs of the two-input two-output optical coupler passes through one of the pulse generators and the other passes through the other of the pulse generators, the input light Differs by an integer multiple of the wavelength,
When the optical path length from the input optical port to the other of the outputs of the two-input two-output optical coupler passes through one of the pulse generators, the input light passes through the other of the pulse generators. It differs by a half integer multiple of the wavelength,
The group delay of the optical path from one of the outputs of the 2-input 2-output optical coupler through one of the data modulators to the optical combiner is reduced from the other of the outputs of the 2-input 2-output optical coupler to the data modulator. It is equal to the group delay of the optical path that passes through the other of the above and reaches the optical multiplexing unit.

The present invention also provides
Each of the pulse generators is a push-pull drive type Mach-Zehnder modulation circuit that is bias-adjusted so that the output light intensity is minimized when not driven.

The present invention also provides
Each of the data modulation units is an optical orthogonal modulation unit or a polarization multiplexed optical orthogonal modulation unit, and the optical multiplexing unit is a 2-input 1-output optical coupler.

The present invention also provides
Each of the two data modulators is a dual light composed of an optical coupler with one input and two outputs and two optical orthogonal modulators optically connected to two outputs of the optical coupler with one input and two outputs, respectively. An orthogonal modulation unit,
The optical multiplexing unit outputs a first two-input one-output optical coupler that multiplexes outputs of one of the two duplex optical quadrature modulation units, and an output of each of the two duplex optical quadrature modulation units. A second two-input one-output optical coupler to be combined, and a polarization beam combining unit that combines the output of the first two-input one-output optical coupler and the output of the second two-input one-output optical coupler. Configured,
A polarization converter is provided between either the output of the first 2-input 1-output optical coupler or the output of the second 2-input 1-output optical coupler and the polarization multiplexer. It is characterized by.

  According to the present invention, it is possible to provide an OTDM optical modulator that has a narrow spectrum width of an output optical signal, can flexibly select a wavelength and a symbol rate, and has a simple configuration.

The block diagram which shows the OTDM modulator which concerns on a 1st reference example . Explanatory drawing which shows operation | movement of an OTDM modulator. The block diagram which shows the OTDM modulator which concerns on a 2nd reference example . The block diagram which shows the OTDM modulator which concerns on a 1st Example . The characteristic view which shows the comparison of an optical signal spectrum. The characteristic view which shows the relationship between an optical signal spectral width and a clock amplitude. The block diagram which shows an optical orthogonal modulation part. The block diagram which shows a polarization multiplexing optical orthogonal modulation part. The block diagram which shows the OTDM modulator which concerns on a 2nd Example . The block diagram which shows the OTDM modulator which concerns on a 3rd Example . The block diagram which shows an example of the conventional OTDM modulator.

The present invention relates to a circuit configuration of an optical modulator, and the effect thereof does not depend on the material forming the modulation circuit. Therefore, the material is not particularly specified in the following reference examples and examples.
As a material for forming the modulation circuit, LiNbO ₃ (LN), KTa _1-x Nb _x O ₃ , K _1-y Li _y Ta ₁ having a Pockels effect which is a kind of electro-optic (EO) effect. _-x Nb _x O ₃ and other multi-component oxide crystals, Pockels effect and Quantum Confined Stark Effect (QCSE) GaAs and InP compound semiconductors capable of modulating the refractive index, EO effects such as chromophore For example, a polymer having
Furthermore, in order to fabricate a modulation circuit having a complicated configuration with low loss, a heterogeneous substrate junction type configuration of the material substrate and a quartz lightwave circuit (PLC) may be used.

The effect of the present invention can be obtained in the same manner regardless of whether the modulation electrode of the Mach-Zehnder modulation unit is a single-ended type or a differential type.
As is generally well known, the arrangement of the modulation electrode in the push-pull drive type Mach-Zehnder modulation circuit depends on the type of substrate, crystal axis direction, and the like. For example, in general, when using an X-cut LN substrate, a single-ended type is used, and when using a Z-cut LN substrate, a differential type is used. can do). Usually, a single-ended signal electrode is disposed at the center of both optical waveguide arms, and a differential signal electrode is disposed immediately above each arm (however, a single-ended electrode using a domain-inverted Z-cut LN substrate) In this case, the electrode is disposed immediately above the arm).

In the reference examples and examples shown below, a single-ended electrode is basically assumed to simplify the drawing, but the response characteristics of the Mach-Zehnder modulator are the same in the case of a differential electrode. As a result, the selection of the electrode arrangement does not affect the effect of the present invention.
In the drawings of the reference examples and examples shown below, only signal electrodes are shown for the sake of simplicity, and ground electrodes are omitted.

In the reference examples and examples shown below, the optical path lengths of both arms of the Mach-Zehnder modulation unit are all designed to be equal. Actually, a deviation in the optical path length occurs due to a process error, a DC drift, or the like. Generally, such a deviation is compensated by adjusting the DC bias. The amount of compensation varies depending on the material, manufacturing conditions, usage environment of the modulator, etc., and is not uniquely determined.
For this reason, in the following reference examples and examples, the value of the inter-arm phase difference provided by the DC bias does not include the optical path length compensation.

( First reference example )
FIG. 1 shows an optical modulator 100 according to a first reference example of the present invention. The optical modulator 100 includes an input optical port 101, a pulse generator 111, a 1-input 2-output optical coupler 131, data modulators 121 and 122, an optical delay unit 141, an optical multiplexer 132, and an output optical port 102. . As the optical coupler 131 and the optical multiplexer 132, a Y-shaped coupler is used.

となる。但しＶ（ｔ）は駆動電圧信号、Ｖπはマッハツェンダ変調回路の半波長電圧である。パルス発生部１１１はクロック源１５１から出力される周波数Ｂ／２（Ｈｚ）のＳｉｎ波信号（直流成分を含まない）によって駆動されるので、駆動電圧信号は、

と表すことができる。ただしＶ_ppは駆動電圧振幅であり、Ｖ_pp＞０である。 As the pulse generator 111, a push-pull drive type Mach-Zehnder modulation circuit is used, and is biased so that the light transmittance is minimized when not driven. If the optical electric field amplitude of the input CW light is E ₀ and the optical frequency component (carrier component) of the input CW light is omitted for _simplicity, the output optical electric field E _MZM (t) of the push-pull drive type Mach-Zehnder modulation circuit under the above bias conditions Is

It becomes. However, V (t) is a drive voltage signal, and Vπ is a half-wave voltage of the Mach-Zehnder modulation circuit. Since the pulse generator 111 is driven by a sine wave signal (not including a DC component) having a frequency B / 2 (Hz) output from the clock source 151, the drive voltage signal is

It can be expressed as. However, V _pp is the drive voltage amplitude, and V _pp > 0.

但しＪ_mは第一種ベッセル関数である。 From the equations (1) and (2), the following equation (3) is obtained.

However, _Jm is a first-type Bessel function.

Since the output signal E _MZM (t) has only a frequency component that is an odd multiple of B / 2 and does not include a carrier component (DC component), as shown in Expression (3), the carrier signal is generally used. -Called asuppressed RZ (CSRZ) pulse. The symbol timing is t = (k + 1/2) / B (k is an integer), and the optical phase of the pulse is different by π between adjacent symbols.

The CSRZ pulse generated by the pulse generator 111 is branched into two by the 1-input 2-output optical coupler 131, modulated by the data modulators 121 and 122, and then multiplexed by the optical multiplexer 132, and output optical port 102. Is output from.
Between one output of the 1-input 2-output optical coupler 131 and the data modulator 122, an optical delay unit 141 that provides a group delay 1 / (2B) is disposed. Therefore, the optical electric field waveform E ₁ (t) and the optical electric field waveform E ₂ (t) immediately before the data modulation units 121 and 122 are expressed by the following equations (4-1) and (4-2).

The coefficient of amplitude is 1/2 ^1/2 of the equation (3), which corresponds to the 1-input 2-output coupler 131. The symbol timing of E ₁ (t) is t = (k + 1/2) / B, and the symbol timing of E ₂ (t) is t = k / B (k is an integer).

As the data modulation units 121 and 122, an optical orthogonal modulation unit 700 shown in FIG. 7 is used. The optical quadrature modulation unit 700 has a general configuration in which Mach-Zehnder modulation circuits 761 and 762 are connected in parallel by optical couplers 731 and 732 and an optical phase adjustment unit 751 that provides an optical phase difference π / 2 between both arms is provided. is there.
As generally used, the output-side optical coupler 732 may have two inputs and two outputs, and one of the outputs may be a main output and the other may be a monitor output. The configuration of the optical quadrature modulation unit 700 is the same as that of a commercially available QPSK modulator, and is well known, so detailed description thereof is omitted.
In FIG. 7, reference numeral 701 denotes an input optical port, and reference numeral 702 denotes an output optical port.

As the data modulation units 121 and 122, a polarization multiplexed optical orthogonal modulation unit 800 shown in FIG. 8 can be used. The polarization multiplexed optical quadrature modulation unit 800 includes an input optical port 801, a 1-input 2-output optical coupler 831, optical orthogonal modulation units 881 and 882 connected to the respective output ports, and one optical orthogonal modulation unit 882. Connected to the output of the polarization beam combiner 835 connected to the output of the polarization converter 853 connected to the subsequent stage, the output of the other optical orthogonal modulator 881 and the output of the polarization converter 853, and the output of the polarization combiner 835 The output optical port 802 is configured.
The optical quadrature modulation units 881 and 882 can have the same configuration as the optical quadrature modulation unit 700 in FIG.
Such a configuration of the polarization multiplexed optical quadrature modulation unit 800 is the same as that of a commercially available polarization multiplexed QPSK modulator, and is well known, and thus detailed description thereof is omitted.

  The drive symbol rate of the data modulation units 121 and 122 is B (sps: symbol / second). Any modulation method can be used. For example, if the Mach-Zehnder modulation circuits 761 and 762 shown in FIG. (Keying, QPSK) can be performed, and 16-level quadrature amplitude modulation (16QAM) of symbol rate B can be performed by driving with a four-value data signal.

  FIG. 2 is a diagram schematically illustrating the operation (upper stage) of the present optical modulator 100 using the CSRZ pulse while comparing it with the operation (lower stage) of the configuration using the conventional RZ pulse.

First, the operation (upper stage) of the present optical modulator 100 will be described. Here, for simplicity, ignoring the higher-order terms in equations (4-1) and (4-2) and considering only the term of n = 0, the output optical field of the pulse generator 111 does not include a DC component. It becomes a Sin waveform of frequency B / 2. The waveform on the arm 2 side where the data modulation unit 122 is arranged is given a group delay of 1 / (2B), that is, a quarter cycle, by the optical delay unit 141 with respect to the waveform on the arm 1 side where the data modulation unit 121 is arranged. It has been. Therefore, comparing the pulse time waveforms immediately before the data modulators 121 and 122 of each arm, the optical electric field intensity | E | ² is the maximum on the arm 1 side at the timing indicated by ▽ in the upper left column of FIG. At the timing indicated by Δ, the optical electric field intensity is maximum on the arm 2 side and zero on the arm 1 side.
For this reason, if the symbol timings in the data modulation units 121 and 122 are matched with ▽ and Δ, respectively, the outputs of the data modulation units 121 and 122 will not interfere with each other even after the outputs of the data modulation units 121 and 122 are combined. Can be obtained. Each of ▽ and Δ appears in a 1 / B cycle, which matches the driving symbol rate B of each data modulation unit 121, 122.

Considering the optical signal spectrum, the CSRZ pulse waveform before data modulation is expressed by time shifts of equations (4-1) and (4-2) if the higher-order components are ignored in both arms. 2 As shown in the upper middle row, the carrier frequency is ± B / 2. Assuming that the drive data signal is an NRZ signal, the CSRZ optical signal spectrum immediately after the data modulators 121 and 122 in the optical modulator 100 is obtained by superimposing the spectrum of the NRZ signal on the spectrum of the CSRZ pulse as shown in the upper right column of FIG. It becomes. Since the spectrum of the NRZ signal has the 1st null point at the carrier frequency ± B, the spectrum width defined by the 1st null point interval of the spectrum of the CSRZ signal is about 3B. Since the spectrum shape does not change even after the signals of arms 1 and 2 are combined, the spectrum width of the final output optical signal is about 3B, which is 1. About 5 times.
Thus, one of the major features of this reference example is that the spectrum width of the final output optical signal can be reduced to about 3B.

  On the other hand, in the conventional method shown in Patent Literature 1, Patent Literature 2, and Non-Patent Literature 3 (FIG. 11), the pulse generator (first optical modulation means 2) is driven by a clock signal having a frequency B. RZ pulse is used. In this case, the optical electric field waveform immediately before each data modulation unit (second and third light modulation means 3 and 4) is a Sin waveform of frequency B including a DC component as shown in the lower left column of FIG. For this reason, the spectrum of the pulse (ignoring higher order components) is composed of a carrier component and a carrier frequency ± B component as shown in the lower middle column of FIG. 2, and the spectrum width of the final output optical signal after data modulation is shown in FIG. 2 As shown in the lower right column, it is about 4B. This is about twice the symbol rate of the final optical output signal.

  Although some variations are shown in the modulator configuration of the conventional method, since the RZ pulse is used for all, the above discussion is valid in common. Further, when a Mach-Zehnder modulation circuit is used as a pulse generator in the conventional method, since it is used under a bias condition different from the CSRZ mode, the equations (1), (2), (3), (4-1), (4-2) ) Does not hold.

The arrangement of the optical delay unit 141 may be between the data modulation unit 122 and the optical multiplexing unit 132. In that case, the pulse waveform immediately before the data modulation unit in FIG. 2 matches between both arms, and the timings of ▽ and Δ overlap in time, but a group delay difference of 1 / (2B) is given after data modulation. 2 is obtained.
Similarly, even if the optical delay unit is divided and arranged before and after the modulation unit 122, an equivalent operation can be obtained if the total group delay is 1 / (2B).

Even if the group delay amount is increased or decreased by an integral multiple of the driving symbol interval 1 / B, the condition that the signals from both arms do not interfere with each other in the symbol timing in the combined OTDM signal is maintained. May be (p + 1/2) / B, that is, a half-integer multiple of 1 / B, where p is an integer. Furthermore, the optical delay unit designed so that the group delay amount is (p + 1/2) / B with respect to the symbol rate B is such that q is an integer, and the following equations 5-1 and 5-2 hold: It can also be used at the symbol rate B '.

As described above, by using the present optical modulator 100, an OTDM signal having a symbol rate 2B that is twice the driving symbol rate B and having a spectral width of 3B, which is 3/4 of the conventional method. (Output optical signal) can be obtained.
The present optical modulator 100 operates by inputting CW light, and its operating characteristics are not dependent on the input light wavelength in principle. The optical modulator 100 includes a general push-pull type Mach-Zehnder modulation circuit and an optical quadrature modulation circuit, and does not require an external pulse shaping device.
Regarding the flexibility for changing the symbol rate, the pulse generator 111 can cope with the change of the symbol rate by changing the drive clock frequency, and the optical delay unit 141 has the conditions shown in the equations (5-1) and (5-2). It is possible to respond to symbol rate changes within the range.

In this reference example , a Y-shaped coupler is generally used as the optical coupler 131 and the optical coupling unit 132 because the operating wavelength is wide and the branching ratio deviation is small. However, couplers of other structures such as directional couplers and multimode interference are used. A (Multimode Interference, MMI) coupler or the like may be used.

( Second reference example )
FIG. 3 shows an optical modulator 300 according to a second reference example of the present invention. The optical modulator 300 includes an input optical port 301, a 1-input 2-output optical coupler 331, pulse generation units 311 and 312, data modulation units 321 and 322, an optical multiplexing unit 332, and an output optical port 302. As the optical coupler 331 and the optical multiplexing unit 332, Y-shaped couplers are used.

Unlike the optical modulator 100 shown in FIG. 1, in this optical modulator 300, pulse generators 311, 312 are connected to the outputs of the 1-input 2-output optical coupler 331, respectively, and data modulators 321, 322 are further downstream. Is connected. In addition, the optical modulator 300 does not include an optical delay unit, and instead, an electrical delay unit 361 is provided between the clock source 351 and the pulse generation unit 312. The clock frequency, the structure and bias condition of the pulse generator, the structure of the data modulator and the drive symbol rate are the same as those in the first reference example .

The clock source 351 generates a clock signal having a clock frequency of B / 2, and this clock signal is directly input to the pulse generator 311 and input to the pulse generator 312 via the electrical delay unit 361.
The electrical delay unit 361 is given a clock phase delay π / 2. Since the clock frequency is B / 2, this clock phase delay is 1 / (2B) in terms of time delay. Optical electric field waveform of the output CSRZ pulses from the for pulse generator 312, since on the time axis to the output from the pulse generating unit 311 1 / (2B) i.e. 1/4 cycle late, first shown in the end view 2 The operation equivalent to that of the first reference example can be obtained.

Since the present optical modulator 300 does not include an optical delay unit, the flexibility in changing the symbol rate is improved as compared with the optical modulator 100 shown in FIG. That is, to change the symbol rate, the operating frequency of the clock source 351 and the delay amount of the electrical delay unit 361 may be changed, and the optical circuit side can in principle cope with any symbol rate (electrical-optical frequency response). (Within a frequency range determined by characteristic limits).
Usually, since electronic circuit boards having different designs are used for different symbol rates, it is an advantage from the viewpoint of flexibility that the design change on the optical modulator side is not required for changing the symbol rate. Other advantages over the conventional method (spectrum = final symbol rate × 1.5, simple configuration, wavelength-independent operation) can be obtained in the same manner as in the first reference example .

  In this optical modulator 300, a group delay of an optical path from one of the outputs of the optical coupler 331 having one input and two outputs to one of the data modulators 321 and 322 to the optical multiplexer 332, and an optical coupler having one input and two outputs. The group delay of the optical path from the other output of 331 to the optical multiplexing unit 332 through the other of the data modulation units 321 and 323 is made equal.

( First embodiment )
FIG. 4 shows an optical modulator 400 according to the first embodiment of the present invention. The optical modulator 400 includes an input optical port 401, a 1-input 2-output optical coupler 431, pulse generation units 411, 412, an optical phase adjustment unit 441, an input-two-output optical coupler 433, data modulation units 421, 422, optical multiplexing. 432 and an output optical port 402. A Y-shaped coupler is used as the optical coupler 431 and the optical multiplexer 432. A directional coupler is used as the 2-input 2-output optical coupler 433. Reference numeral 451 denotes a clock source, and reference numeral 461 denotes an electrical delay unit.

Unlike the optical modulator 300 shown in FIG. 3, in the present optical modulator 400, the outputs of the pulse generators 411 and 412 are connected to two inputs of a two-input two-output optical coupler 433, respectively, and a two-input two-output optical coupler. The two outputs of 433 are connected to the data modulators 421 and 422, and an optical phase adjustment unit 441 is provided between the pulse generation unit 412 and the 2-input 2-output optical coupler 433. Other conditions, that is, the clock frequency, the structure and bias condition of the pulse generator, the structure of the data modulator and the drive symbol rate are the same as those in the first reference example and the second reference example .

In the optical modulator 400 of the present example, CSRZ pulses shifted from each other by ¼ period are output from the outputs of the 2-input 2-output optical coupler 433, and are substantially equivalent to the operation of the first reference example shown in FIG. You can get action. First, this is shown by a mathematical formula.

高次（ｎ≧１）の項を無視すれば、式（４−１），（４−２）の場合と同様、Ｅ₁（ｔ）およびＥ₂（ｔ）は互いに１／４周期異なる周波数Ｂ／２のＳｉｎ波形となっていることがわかる。Ｅ₁（ｔ）のシンボルタイミングはｔ＝（ｋ＋３／４）／Ｂ、Ｅ₂（ｔ）のシンボルタイミングはｔ＝（ｋ＋１／４）／Ｂ（ｋは整数）である。 The clock phase delay provided by the electrical delay unit 461 is π / 2, and this delay corresponds to the time delay 1 / (2B).
The optical path length from the input optical port 401 to the output on the arm 2 side of the 2-input 2-output optical coupler 433 is an integral multiple of the input wavelength when passing through the pulse generator 411 and when passing through the pulse generator 412 ( The optical phase difference is adjusted to be an integral multiple of 2π (including zero). At this time, the optical path length from the input optical port 401 to the output on the arm 1 side of the 2-input 2-output optical coupler 433 is a half integer multiple of the input wavelength when passing through the pulse generator 411 and when passing through the pulse generator 412. (Ie, the optical phase difference is an odd multiple of π). Therefore, the optical electric field waveforms E ₁ (t) and E ₂ (t) immediately before the data modulation units 411 and 412 can be expressed by the following equations, respectively.

If the higher-order (n ≧ 1) terms are ignored, E ₁ (t) and E ₂ (t) are different from each other by a quarter period, as in the case of equations (4-1) and (4-2). It can be seen that the S / 2 waveform is B / 2. The symbol timing of E ₁ (t) is t = (k + 3/4) / B, and the symbol timing of E ₂ (t) is t = (k + 1/4) / B (k is an integer).

Similar to the second reference example shown in FIG. 3, the present embodiment does not include an optical delay unit, and therefore can flexibly cope with a change in symbol rate. Other advantages over the conventional method (spectrum = final symbol rate × 1.5, simple configuration, wavelength-independent operation) can be obtained in the same manner as in the first reference example .

Furthermore, this example has an advantage that the side lobe intensity of the output optical signal spectrum can be suppressed as compared with the first reference example and the second reference example . This is because the relative phase of the harmonic (n ≧ 1) relative to the fundamental wave (n = 0 term) is different between the equations (6-1), (6-2) and (3). ing. By suppressing the side lobe intensity, it is possible to suppress crosstalk between adjacent wavelength channels during wavelength multiplexing transmission.

で表されるものと仮定した。但しｃ_1,kおよびｃ_2,kは夫々データ変調部４１１，４１２によって与えられる複素シンボル値であり、例えばＱＰＳＫであれば±１±ｊの４値のいずれかをとる。シンボル列｛ｃ_1,k｝と｛ｃ_2,k｝は互いに相関しない。また、ｔ₀はシンボルタイミングを表すパラメータであり、Ｅ₁（ｔ）とＥ₂（ｔ）が式（３）で表される第１の参考例および第２の参考例においてはｔ₀＝１／（２Ｂ）、式（６−１），（６−２）で表される第１の実施例においてはｔ₀＝３／（４Ｂ）である。 FIG. 5 shows the output optical signal spectrum (dotted line) in the second reference example shown in FIG. 3 and the output optical signal spectrum (solid line) in the first embodiment shown in FIG. These spectra are theoretical curves, the horizontal axis is the normalized frequency (relative optical frequency relative to the input optical frequency from the modulator divided by the drive symbol rate B), and the vertical axis is the normalized intensity (peak value is 0 dB). Displayed in dB scale). Since the spectrum shape changes depending on Vpp, the cases where Vpp is 0.5, 1.0, 1.5, and 2.0 times Vπ are shown. Although the spectrum shape depends on the response waveform of the data modulation unit, it is assumed here that the response waveform of the data modulation unit is a square cosine pulse waveform for simplicity. That is, the optical electric field waveform E _out (t) finally output from the output port 402 is

It is assumed that However, c _{1, k} and c _{2, k} are complex symbol values given by the data modulation units 411 and 412, and take one of four values of ± 1 ± j in the case of QPSK, for example. The symbol sequences {c _{1, k} } and {c _{2, k} } are not correlated with each other. T ₀ is a parameter representing symbol timing, and t ₁ = 1 in the first reference example and the second reference example in which E ₁ (t) and E ₂ (t) are expressed by Expression (3). In the first embodiment represented by / (2B) and formulas (6-1) and (6-2), t ₀ = 3 / (4B).

In FIG. 5, it can be seen that although the solid line spectrum is slightly wider than the dotted line spectrum, the side lobe intensity appearing at the normalized frequency> 1.5 is suppressed compared to the dotted line spectrum. However, when Vpp is 1.5 Vπ or more, the 1st null point near the normalized frequency 1.5 is not clearly observed.
For this reason, the definition of the spectrum width is defined by the full width of −20 dB with respect to the peak intensity, and plotted against Vpp / Vπ, FIG. 6 is obtained. As shown in FIG. 6, in the case of the second reference example (dotted line) shown in FIG. 3, the spectrum width gradually decreases as Vpp increases, whereas in the case of the first embodiment shown in FIG. In (solid line), the spectrum width increases as Vpp increases. It can be seen that in order to obtain a spectrum width equal to or less than three times (3B) the drive symbol rate as a guide, in this example, Vpp <1.8 Vπ may be set.

  In this embodiment, a low-loss directional coupler is generally used as the 2-input 2-output optical coupler 433. However, a 2-input 2-output optical coupler having another structure, such as a 2-input 2-output MMI coupler, may be used. good. An MMI coupler generally has a slightly larger loss than a directional coupler, but has a smaller wavelength dependency.

  Further, in the present optical modulator 400, a group delay of an optical path from one of the outputs of the optical coupler 433 having two inputs and two outputs to one of the data modulators 421 and 422 to the optical multiplexer 432, and an optical coupler having two inputs and two outputs. The group delay of the optical path from the other output of 433 to the optical multiplexing unit 432 through the other of the data modulation units 421 and 422 is made equal.

( Second embodiment )
FIG. 9 shows an optical modulator 900 according to the second embodiment of the present invention. The optical modulator 900 includes an input optical port 901, a 1-input 2-output optical coupler 931, pulse generators 911, 912, data modulators 921, 922, an optical multiplexer 932, and an output optical port 902. The pulse generators 911 and 912 are driven by a clock having a frequency B / 2 output from the clock source 951, and a clock phase delay π / 2 is given to the drive clock of the pulse generator 912 by the electric delay unit 961. The configurations and driving conditions of the optical coupler 931 and the pulse generators 911 and 912 are the same as those of the second reference example shown in FIG.

  As the data modulators 921 and 922, a duplex optical quadrature modulator is used. The data modulation unit 921 includes a 1-input 2-output optical coupler 935 and optical orthogonal modulation units 981 and 982 connected to the respective outputs. The data modulation unit 922 includes a 1-input 2-output optical coupler 936 and optical orthogonal modulation units 983, 984 connected to the respective outputs. As the optical orthogonal modulation units 981 to 984, a configuration equivalent to the configuration shown in FIG. 7 can be used. Each of the optical orthogonal modulation units 981 to 984 is driven at a symbol rate B.

  The optical multiplexing unit 932 includes a two-input one-output optical coupler 937 connected to the outputs of the optical quadrature modulation units 981 and 983, and a two-input one-output optical coupler 938 connected to the outputs of the optical quadrature modulation units 982 and 984 and two inputs. A polarization converter 952 connected to the output of the one-output optical coupler 938, a two-input one-output optical coupler 937, and a polarization beam combiner 934 connected to the output of the polarization converter 952.

The function of this configuration is basically the same as the configuration in which the polarization multiplexed optical orthogonal modulation unit 800 shown in FIG. 8 is used as the data modulation units 321 and 322 in the second reference example shown in FIG. In the second reference example , the configuration using the polarization multiplexing optical orthogonal modulation unit requires two polarization conversion units and two polarization combining units as a whole, whereas in this example, only one unit is required. The configuration is somewhat simple.

( Third embodiment )
FIG. 10 shows an optical modulator 1000 according to a third embodiment of the present invention. In this example, the spatial arrangement of the optical orthogonal modulation units 1081 to 1084 constituting the duplex optical orthogonal modulation unit used as the data modulation units 1021 and 1022 is nested in the configuration of the second embodiment shown in FIG. The functions are the same as those of the second embodiment shown in FIG.

  That is, the optical modulator 1000 includes an input optical port 1001, a 1-input 2-output optical coupler 1031, pulse generation units 1011 and 1012, data modulation units 1021 and 1022, an optical multiplexing unit 1032, and an output optical port 1002. The pulse generators 1011 and 1012 are driven by a clock having a frequency B / 2 output from the clock source 1051, and the clock phase delay π / 2 is given to the drive clock of the pulse generator 1012 by the electric delay unit 1061.

  As the data modulators 1021 and 1022, dual optical quadrature modulators are used. The data modulation unit 1021 includes a 1-input 2-output optical coupler 1035 and optical orthogonal modulation units 1081 and 1083 connected to the respective outputs. The data modulation unit 1022 includes a 1-input 2-output optical coupler 1036 and optical orthogonal modulation units 1082 and 1084 connected to the respective outputs.

  The optical multiplexing unit 1032 includes a two-input one-output optical coupler 1037 connected to the outputs of the optical quadrature modulation units 1081 and 1082, and a two-input one-output optical coupler 1038 connected to the outputs of the optical quadrature modulation units 1083 and 1084. A polarization converter 1052 connected to the output of the one-output optical coupler 1038, a two-input one-output optical coupler 1037, and a polarization beam combiner 1034 connected to the output of the polarization converter 1052.

In the present embodiment and the second embodiment, since the second reference example and equivalent arrangements showing the portion of the structure except for the data modulation unit and the optical multiplexing section 3, obtained effect of the second It is equivalent to the reference example .
Naturally, in this example and the second embodiment , if the configuration of the part excluding the data modulation unit and the optical multiplexing unit is the same as the configuration of the first reference example shown in FIG. 1, it is the same as the first reference example. If the same part is made to have the same configuration as that of the first embodiment shown in FIG. 4, the same effect as that of the first embodiment can be obtained.
In addition to the sin waveform signal, the “periodic waveform signal” includes a periodic square waveform signal and a periodic triangular waveform signal.

  The present invention is applicable to an OTDM optical modulator.

100, 300, 400, 900, 1000 Optical modulator 101, 301, 401, 901, 1001 Input optical port 102, 302, 402, 902, 1002 Output optical port 111, 311, 312, 411, 412, 911, 912, 1011, 1012 Pulse generator 121, 122, 321, 322, 421, 422, 921, 922, 1021, 1022 Data modulator 131, 331, 431, 433, 931, 1031 Optical coupler 132, 332, 432, 932 DESCRIPTION OF SYMBOLS 1032 Optical multiplexing part 141 Optical delay part 151,351,451,951,1051 Clock source 361,461,961,1061 Electrical delay part 441 Optical phase adjustment part 700 Optical orthogonal modulation part 800 Polarization multiplexing optical orthogonal modulation part

Claims (4)

One optical port for input,
A 1-input 2-output optical coupler optically connected to the one input optical port;
Two pulse generators optically connected to two outputs of the one-input two-output optical coupler;
A two-input two-output optical coupler optically connected to respective outputs of the two pulse generators;
Two data modulators optically connected to the two outputs of the two-input two-output optical coupler;
An optical multiplexing unit optically connected to the respective outputs of the two data modulation units,
The two data modulators are each driven by a data signal with a symbol rate B, and the two pulse generators are respectively driven by signals having a periodic waveform with a frequency B / 2 having a phase different from each other by π / 2,
When the optical path length from the input optical port to one of the outputs of the two-input two-output optical coupler passes through one of the pulse generators and the other passes through the other of the pulse generators, the input light Differs by an integer multiple of the wavelength,
When the optical path length from the input optical port to the other of the outputs of the two-input two-output optical coupler passes through one of the pulse generators, the input light passes through the other of the pulse generators. It differs by a half integer multiple of the wavelength,
The group delay of the optical path from one of the outputs of the 2-input 2-output optical coupler through one of the data modulators to the optical combiner is reduced from the other of the outputs of the 2-input 2-output optical coupler to the data modulator. An optical modulator characterized by being equal to a group delay of an optical path that passes through the other of the optical paths to the optical multiplexing unit. 2. The optical modulator according to claim 1, wherein each of the pulse generators is a push-pull drive type Mach-Zehnder modulation circuit that is bias-adjusted so that the output light intensity is minimized when not driven. Each of the data modulation unit, an optical quadrature modulator or polarization multiplexed light quadrature modulation unit, the optical modulator according to claim 1, wherein the optical multiplexing unit is two inputs and one output optical coupler . Each of the two data modulators is a dual light composed of an optical coupler with one input and two outputs and two optical orthogonal modulators optically connected to two outputs of the optical coupler with one input and two outputs, respectively. An orthogonal modulation unit,
The optical multiplexing unit outputs a first two-input one-output optical coupler that multiplexes outputs of one of the two duplex optical quadrature modulation units, and an output of each of the two duplex optical quadrature modulation units. A second two-input one-output optical coupler to be combined, and a polarization beam combining unit that combines the output of the first two-input one-output optical coupler and the output of the second two-input one-output optical coupler. Configured,
A polarization converter is provided between either the output of the first 2-input 1-output optical coupler or the output of the second 2-input 1-output optical coupler and the polarization multiplexer. The optical modulator according to claim 1 .

Images