JP3270626B2 - Direct polarization modulation light source control method and integrated semiconductor device driving method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for controlling a light source used for direct polarization modulation and a method for driving an integrated semiconductor device .

[0002]

2. Description of the Related Art In recent years, as information has become multimedia,
There is an increasing demand for a large-capacity communication system regardless of a subscriber system or a LAN. In the wavelength division multiplexing communication, a plurality of independent lines can be provided in one optical fiber which is a transmission line, and a flexible and large-capacity communication system can be provided in combination with the inherent high speed of the optical communication.

In wavelength division multiplexing communication, WDM (Wavelength Division) is used depending on the interval between multiplexed wavelengths.
Multiplexing and optical FDM (Freque)
ncy Division Multiplexin
g). While the wavelength interval of WDM is 1 nm or more, the wavelength interval of optical FDM is as narrow as about 0.1 nm (about 10 GHz at an optical frequency of 1.55 μm band), so that a multiplex system of several tens or more is realized. It is possible. In the optical FDM, the amount of a wavelength difference such as a wavelength interval can be expressed by an optical frequency. Therefore, in the following description, the amount of the wavelength difference is expressed by an optical frequency.

In the optical FDM, since the optical frequency interval is narrow, it is necessary to narrow the occupied optical frequency band of the light source of each channel. Currently, semiconductor lasers are generally used as light sources for optical communication. The occupied optical frequency band of the semiconductor laser is determined by low frequency drift and high frequency chirping. In order to suppress the drift, it is effective to perform feedback control using a wavelength reference on the semiconductor laser. Techniques for suppressing drift to several tens of MHz or less have reached a practical level.

In order to suppress chirping, it is necessary to devise a modulation method. A typical method is direct FSK
(Frequency Shift Keying) modulation method, external intensity modulation method, direct polarization modulation method (Japanese Unexamined Patent Publication No. 2-1597981) and the like have been devised.

The external intensity modulation system includes a system using an external modulator completely independent of a semiconductor laser as a light source, and a system using an external modulator integrated with a semiconductor laser. In the former, modulation without chirp is possible, but the configuration is complicated. On the other hand, the latter has a chirp of several tens GHz. The direct FSK modulation method uses DFB (Distributed FSK).
This can be realized by biasing an EED (Laser Diode) above its threshold value to correspond to the mark or space of the modulation signal and modulating the DFB-LD with a current of about several mA. Since the difference between the optical frequencies of the mark and the space is set to several GHz, the occupied optical frequency width does not become smaller than this optical frequency difference. The direct polarization modulation method uses a polarizing plate and an LD having a characteristic that the polarization direction of output light switches to TE or TM according to the amount of current to be injected. The direction of the polarization of the LD output is switched in accordance with the modulation mark and space, and only one of the polarizations is extracted by a polarizing plate to output an intensity-modulated signal to the outside.

[0007] Of the above three systems, the direct polarization modulation system has an overall advantage over the other two systems in terms of simplicity of configuration, narrower occupied optical frequency width, and the like. Hereinafter, this will be described. FIG. 19 shows the configuration of the direct polarization modulation system.
As shown, the two-electrode DFB-LD 11-1, the adder 15-1, the drive circuits 11-10-1, 11-10-2,
It is composed of polarizers 11-11. 2-electrode DFB-L
D11-1 can switch the polarization of oscillation light by changing the injection current of the two electrodes. Such characteristics can be realized by adjusting device parameters such as the Bragg wavelength of the diffraction grating and the gain spectrum of the active layer. Driving circuits 11-10-1 and 11-10
-2 outputs a current corresponding to the input signal. The current is injected into two electrodes of the two-electrode DFB-LD 11-1. The adder 15-1 adds two input signals (a bias signal and a modulation signal). The output of the adder 15-1 is connected to the drive circuit 11-10-1.
A bias signal is directly input to 0-2. The modulation signal and the two bias signals are input from a transmitter in which the light source is incorporated. Polarizer 11-11 is a two-electrode DFB-
Only the TE polarization of the output light of the LD 11-1 is transmitted, and the modulation output is sent to the transmission line.

FIG. 20 shows a two-electrode DFB-L for direct polarization modulation.
It is a sectional perspective view of D. In the figure, 16-1 is an n-InP substrate, 16-2 is an n-InP buffer layer on which a diffraction grating is formed, 16-3 is an n-InGaAsP lower light guide layer,
16-4 is an active layer having a strained superlattice structure of i-InGaAsP, 16-5 is a p-InP cladding layer, and 16-6 is a p-
InGaAsP contact layer, 16-7 is high-resistance InP
A buried layer, 16-8 is an electrode isolation region from which the contact layer 16-6 has been removed, 16-9 is a Cr / AuZnNi / Au layer which is a front side (emission side) electrode, and 16-10 is a Cr / Au layer which is a rear side electrode. An AuZnNi / Au layer, 16-11 is an AuGeNi / Au layer serving as a substrate-side electrode, and 16-12 is a SiO film serving as an antireflection film. This element has a multiple quantum well structure in which the active layer has a tensile strain, has a lower TM polarization oscillation threshold value than a normal DFB-LD, and can efficiently perform polarization modulation.

FIG. 21 is a schematic diagram showing the oscillation characteristics of a two-electrode DFB-LD having switching characteristics of TE polarization and TM polarization. The injection current I ₁ of the front electrode on the horizontal axis, the rear placed vertically injection current I ₂ of the electrode, TE (right lower than the thick dotted line) area oscillates in polarization and region (thick dotted line that oscillates in the TM polarization (Upper left). The curves in each region indicate contour lines of the output of each polarization, and the output increases as going inward.

FIG. 22 is a schematic diagram showing a state of switching between TE polarization and TM polarization. The injection current of the front electrode is I
Shows the change in the light intensity of the polarization of each of the case of fixed to _1b, changing the injection current I ₂ of the rear electrode. The vicinity of the switching region (described below) is enlarged. I ₂ <
In I _2smin , it oscillates only with TE polarization, I _2smin <I ₂ <I
_{At 2smax} , TE polarization and TM polarization coexist, and when I ₂ > I _2smax , oscillation occurs only with TM polarization. When I _2smin <I ₂ <I _2smax , the oscillation state becomes unstable in both the TE polarization and the TM polarization, and as the time average of the light intensity increases, I ₂
E polarization decreases and TM polarization increases. I _2sc is a point where the light intensity of the TE polarization and the TM polarization becomes equal. Also, an area I _2Smax from I _2Smin in the following description referred to as a switching region.

[0011] Direct polarization modulation can be realized by setting a bias point in the switching region and performing modulation. Here, a case where I ₁ is fixed and I ₂ is modulated will be described (refer to FIGS. 21 and 22 in the following description). The I ₁ is fixed to the I _1b. TE polarization at this time and TM
Polarization switching is performed by _changing I ₂ from I _2smin or less to I _2
Occurs when changing to a value of _2smax or more. I ₂ of the bias component I _2b, a modulation component _{I 2m, I 2b <I 2smin} , I 2b + I
_2m > _I2smax . This gives I ₂ =
TE polarization when I _2b , T when I ₂ = I _2b + I _2m
Oscillates with M polarization. By extracting only the TE polarization of this optical signal by the polarizer 11-11, an intensity-modulated optical signal can be produced. In this example,
When I ₂ = I _2b , the light output is ON, and when I ₂ = I _2b + I _2m , the light output is OFF, so that the modulated signal is inverted by this light source.

[0012] Since the structure of this modulation system directly modulates the DFB-LD, the structure is as simple as that of the direct FSK modulation system. The amplitude of the modulation current is as small as several mA, and the modulation signal mark is small. Since the light is emitted in either of the space and the space, chirping of the optical frequency of the optical signal obtained by the modulation is as small as that of the external modulation method using an external modulator independent of the semiconductor laser.

Incidentally, the direct polarization modulation can be expressed in such a manner that it can be realized by setting a bias point of an injection current in a switching region and performing modulation of an appropriate amplitude. FIG.
An example will be described using 1, 24, and 25. I ₁ is I
_1b , the bias component of I ₂ is I _2b , and the modulation component is a rectangular wave of amplitude I _mod . This square wave has I _{mod / 2 at the} top,
The lower end is -I _{mod / 2} , which corresponds to the modulation mark and space, respectively. At this time, I _2E = I _2b −I _{mod / 2}
Then, I _2b and I _mod are set so that only the TE polarization oscillates, and if I _2M = I _2b + I _{mod / 2} , only the TE polarization oscillates (FIGS. 25A and 25B). By extracting only the TE (or TM) polarization of this optical signal by the polarizer 11-11, an intensity-modulated optical signal can be created (FIG. 25C). In the following description, when referring to a bias component or a bias point, the above two meanings (a case where a modulation component is considered as a DC component and a case where a modulation component is considered as an AC component) are properly used.

[0014]

However, this direct polarization modulation method has the following problems. The distribution of the TE polarization and TM polarization oscillation regions of the multi-electrode (including the two electrodes described above) DFB-LD differs depending on the element. For this reason, the bias point of direct polarization modulation must be set based on precise measurement for each element. Further, the distribution of the TE-polarized light and the TM-polarized light oscillation region varies depending on the temperature or the like even in the same element. For this reason, it is difficult to maintain the state of the polarization modulation (the intensity ratio between the TE polarization and the TM polarization corresponding to the modulation signal) at the best.

Hereinafter, deterioration of the modulation state due to a change in the state of the element will be described in detail with reference to FIGS.
Here, FIG. 23 shows the optical output when the bias component of the injection current deviates from the switching point.

[0016] In FIG. 22, the bias component I _2b of I _2, the initial setting of the modulation component _{I 2m, I 2b <I 2smin} -
δ, I _2b + I _2m > I _2smax + δ (δ is a very small amount).
In this state, the two-electrode DFB-LD 11-1 has I ₂ = I
_2b oscillates with TE polarization, and I ₂ = I _2b + I _2m oscillates with TM polarization. At this time, the light intensity of the optical signal output through the polarizer 11-11 changes as shown in FIG. Since this light source inverts the modulation signal, the intensity of the optical signal is I
Greater when ₂ = I _2b, smaller when _{I 2 = I 2b + I 2m} .

The current value (I _2smin , I 2
_2smax , I _2sc ) and I _2b becomes _greater than I _2smin , the light intensity of the TE polarization when I ₂ = I _2b decreases, and T
The light intensity of M polarization increases. Further, when I _2b becomes larger than I _2smax , only TM polarization is output. The situation at this time is shown in FIG. On the other hand, I _2b
When _2smax -I _2m becomes smaller than, I ₂ = I _2b + light intensity of the TM polarization when the I _2m is reduced, the light intensity of the TE polarization wave is increased. When I _2b is smaller than I _2smin −I _2m , TE
Only the polarization is output. Figure 2 shows the situation at this time.
This is shown in FIG.

As can be understood from the above description, when the current value in the switching region is shifted, the modulation efficiency of the polarization modulation is deteriorated (FIG. 23C), and in some cases, the modulation cannot be performed correctly (FIG. 23). (B)).

Further, in the above-mentioned feedback control method, when using as a light source for optical communication using an optical device such as a polarizing beam splitter or a photodiode other than the semiconductor laser device, it is difficult to make the device compact and modular. In addition, there is a problem that the optical adjustment is required when the module is formed, and the productivity is deteriorated.

Therefore, an object of the present invention is to provide a method for controlling a light source used for direct polarization modulation,
And a method for driving an integrated semiconductor device .

[0021]

According to a first direct-polarization modulation light source control method of the present invention for solving the above-mentioned problems, two orthogonal polarizations of output light of a direct-polarization modulation light source (for example,
Means for separating TE polarized light and TM polarized light), means for receiving the two polarized waves, and means for differentially amplifying the two signals, and one of the polarized lights of the output light from the direct polarized light source. , The intensity difference of the other polarization (each polarization is usually modulated, so the difference in intensity is interpreted in a variety of ways, such as the difference between the average intensity and the difference between the peak intensities. The driving amount of the polarization-modulated light source is directly feedback-controlled so that the value becomes zero or a predetermined value.

[0022]

Thus, at least the modulation bias point can always be maintained in the switching region of one (TE) polarization oscillation region and the other (TM) polarization oscillation region .

More specifically, the following configuration is also possible. Means for differentially amplifying two signals obtained from the means for receiving the two orthogonal polarizations, a reference voltage source,
Means for differentially amplifying a signal obtained from one of the means for receiving the polarized light and a signal from the reference voltage source, such that the intensity difference between the two orthogonal polarizations becomes a predetermined value. Feedback control is performed, and at the same time, feedback control is performed so that the output intensity of one of the two polarized waves becomes constant. A reference voltage source for each of the two orthogonal polarizations, and a unit for differentially amplifying a signal obtained from a unit for receiving the polarization and a signal from the reference voltage source, respectively, Feedback control is performed so that the output intensity of each wave becomes constant. Multi-electrode DFB-L as direct polarization modulation light source
D is used. The means for separating the two orthogonal polarizations is constituted by a polarization beam splitter or means for separating the two orthogonal polarizations by wavelength.

[0025]

[0026]

Further, the integrated optical semiconductor of the second configuration of the present invention
The driving method of the device is to flow a part of the optical waveguide including the light emitting layer.
By modulating the current, two orthogonal polarization modes are switched.
Semiconductor laser having a switching structure and modulation of the semiconductor laser
Output light intensity of the two orthogonal polarization modes of the output light
And means for electrically detecting the
Integrated optical semiconductor device, emitted from the semiconductor device to the outside
In order to stabilize the modulation state of the
To take the difference between electrical signals converted from the output light intensity in wave mode
Stage, a reference voltage source, the output of the means for taking the difference and the reference voltage
Means for differentially amplifying the output of the pressure source and the polarization plane
Semiconductor so that the intensity difference between two polarized waves with different
It is characterized by performing feedback control on the laser.

[0028]

[0029]

[0030]

[0031]

The above configurations and specific means can be combined in any manner possible.

[0033]

Embodiment 1 Hereinafter, a first embodiment of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram of a first embodiment of a direct polarization modulation light source to which the present invention is applied. As shown, the first embodiment has a two-electrode DFB-LD1-1,
Polarization beam splitter 1-2, light receiving element 1-3-1, 1
-3-2, amplifiers 1-4-1, 1-4-2, low-pass filter (hereinafter referred to as LPF) 1-5-1, 1-5
2. Differential amplifier 1-6, feedback control circuit 1-7, switch 1-8, adder 1-9, drive circuit 1-10-1, 1-1
0-2 and a polarizer 1-11 as main parts.

The two-electrode DFB-LD1-1 is the same as the proposed example, and has the structure shown in FIG. 20 and the characteristics shown in FIG. The two-electrode DFB-LD1-1 emits light from two front and rear end faces. The light emitted from the front end face is a polarizer 1-1.
1 extracts only the TE polarization. The light emitted from the rear end face is used for bias point control. First, the light emitted from the rear end face is subjected to TE polarization and T polarization by the polarization beam splitter 1-2.
The M polarization is separated. The TE polarized light among the separated polarized lights is converted into an electric signal by the light receiving element 1-3-1. The electric signal is amplified by the amplifier 1-4-1. The low frequency component is extracted by the LPF 1-5-1. After that, it is input to the plus input terminal of the differential amplifier 1-6. The TM polarization is similarly processed by the light receiving element 1-3-2, the amplifier 1-4-2, and the LPF 1-5-1.
The signal is input to the minus input terminal of the differential amplifier 1-6. The differential amplifier 1-6 amplifies the difference between the two input signals.
The output of the LPF 1-5-1 is the LPF 1-5
When the output of the LPF 1-5-1 is smaller than the output of the LPF 1-5-2, the output becomes negative. The output of the differential amplifier 1-6 becomes an error signal in the present control system.

The feedback control circuit 1-7 includes a differential amplifier 1-6.
Is used as an error signal to generate a control signal. As a feedback control method, a well-known PID (Proportion
nal Integral Differentia
l) Control is preferably used. This feedback control circuit 1-7
Is a control signal in this control system. This control signal is input to the adder 1-9 via the switch 1-8. The switch 1-8 is turned on / off by a control ON / OFF signal from the transmission unit. The bias signal, the modulation signal, and the control signal from the transmission unit are input to the adder 1-9. The adder 1-9 adds these three signals, and the output is input to the drive circuit 1-10-1. The drive circuit 1-10-1 outputs a current corresponding to an input signal to this, and this current is injected into the rear electrode of the two-electrode DFB-LD1-1.

The drive circuit 1-10-2 outputs a current corresponding to the bias signal input from the transmission unit, and the current is injected into the front electrode of the two-electrode DFB-LD1-1.
In the above description, the transmission unit refers to a device in which a control system to which the control method of the present embodiment is applied is incorporated.

FIG. 2 shows the relationship between the bias signal I _{2b of the} current injected into the rear electrode of the two-electrode DFB-LD1-1 and the error signal. The vicinity of the switching region (see FIG. 22) is enlarged.

Next, the operation of this embodiment will be described. Note that the bias to the switching region and the polarization modulation are the same as those in the already proposed example, and therefore will not be described here.

The cutoff frequencies of the LPFs 1-5-1 and 1-5-2 are set sufficiently lower than the modulation signal. Therefore, when the modulation state is good (the case where the TE polarization and the TM polarization are switched in accordance with the modulation signal, and the optical output of the light source is as shown in FIG. 23A), the LPF1-
Signals are obtained from both 5-1 and LPF1-5-2.
I _2b + I _2m / 2 is the center of the switching region I _2sc (FIG. 2)
2), the magnitudes of these two signals are substantially equal (however, in this case, the duty ratio of the modulation signal is 50%).

Oscillation characteristics change due to changes in the temperature and the like of the two-electrode DFB-LD1-1, and the current value (I
_2sc , I _2smin , I _2smax ) degrades the modulation state. When both the relations of I _2b <I _2smin , I _2b > I _2smax −I _2m are satisfied, the deterioration of the modulation state is not so large, but if not, FIGS. c)
As shown in FIG. At this time, LP
The difference between the output signals of F1-5-1 and LPF1-5-2 (the output of the differential amplifier 1-6) increases.

Output signal of differential amplifier 1-6 (error signal)
Is a zero cross signal which takes zero at I _2sc −I _2m / 2 as shown in FIG. Therefore, by feedback-controlling this signal as an error signal, I _2b + I _2m / 2
_Can be adjusted to I _2sc . Thereby, the modulation state can be maintained in a good state as shown in FIG. This feedback control is started by turning on the switch 1-8 after setting I _1b and I _2b as switching points. Further, the deviation of I _1b and I _{2b from} the switching point at the time of setting can be automatically corrected by this feedback control, and the tolerance for this setting increases.

[0042]

Embodiment 2 A second embodiment of the present invention will be described with reference to the drawings. FIG. 3 is a block diagram of a second embodiment of the direct polarization modulation light source to which the present invention is applied. The difference from the first embodiment shown in FIG. 1 is that the outputs of LPF 1-5-1 and LPF 1-5-2 are input to a subtractor 3-1 and the output of the subtracter 3-1 and a reference voltage source 3-2. Is input to the differential amplifier 1-6.

The subtractor 3-1 outputs a value obtained by subtracting the output signal of the LPF 1-5-2 from the output signal of the LPF 1-5-1. The output signal of the subtractor 3-1 is input to the plus input terminal of the differential amplifier 1-6, and the output signal of the reference voltage source 3-2 is input to the minus input terminal of the differential amplifier 1-6.

When the duty ratio of the modulation signal deviates from 50%, the magnitudes of the output signal of LPF 1-5-1 and the output signal of LPF 1-5-2 are not equal in a good modulation state. Accordingly, in this case, in the configuration of the first embodiment, the feedback control stabilizes in a state deviated from a favorable modulation state. The present embodiment improves this.
LPF1-5-1 output signal and LP in good modulation state
The reference voltage source 3 is added to the difference between the magnitudes of the output signals of F1-5-2.
By setting the output of -2, a good modulation state can be maintained even when a modulation signal whose duty ratio deviates from 50% is used. Other operations etc. are first
This is substantially the same as the embodiment.

[0045]

Embodiment 3 In the above embodiment, the two-electrode DFB-LD1-
Feedback control was performed on the injection current of the rear electrode of No. 1,
It is also possible to apply to the injection current of the front electrode. Means for generating an error signal, such as an amplifier, an LPF, and a differential amplifier, is a modulation center value (in the embodiment, I _2b + I _2m
Another means for generating a zero-cross signal which becomes zero when (/ 2) comes to the center I _{2sc of the} switching region may be used.
Although the two-electrode DFB-LD is used as the direct polarization modulation light source, the present invention can be applied to a three-electrode DFB-LD. Further, the present invention can be applied to another light source having a similar function.

[0046]

Embodiment 4 In the case where the shape of the oscillation characteristics of both polarizations changes, the following problem also arises. For example, if the oscillation characteristics of both polarizations change as shown in FIG. 26 (the shapes of the oscillation characteristics of both polarizations are not symmetric), I ₂ = I
_{At 2E} (point B in the figure), the TE polarization intensity is P _TE , I ₂ = I
_{At 2M} (point A in the figure), the intensity of the TM polarization is _PTM , and the intensity of both polarizations is different ( _PTE ≠ _PTM ). At this time, output signals of both polarizations are as shown in FIG. The change in the oscillation characteristics due to such a change in the environment may have the following adverse effects. 1) The modulation efficiency fluctuates due to the fluctuation of the output light intensity. 2) If the carrier density at point A differs from that at point B, chirp occurs in the optical frequency. 3) Polarization switching does not occur, and modulation cannot be performed correctly.

In view of this, a configuration is conceivable in which the bias component of I ₂ is controlled to maintain the bias point of the modulation within the switching region, and that the TE polarization and the TM polarization have the same intensity.

In FIG. 26, the bias component of I ₂ is I _2b ′
At this time, the output light intensities of the TE polarization and the TM polarization are equal. I ₂
If a small deviation from the bias component I _2b ′ (current value at which the output light intensity of the TE polarization and the TM polarization is equal) is δI _2b , the relationship between δI _2b and the error signal is as shown in FIG. (However, FIG. 2 shows the relationship between the bias component and the error signal when the modulation component is considered as a DC component). By inputting this error signal to the feedback control circuit and performing, for example, PID control, δI _2b can always be kept at zero, in other words, the output light intensities of TE polarization and TM polarization can always be kept equal. This control method is also effective depending on conditions, but has the following limitations. Simply controlling the bias point to equalize the output light intensities of both polarizations at all times does not guarantee that the output light intensity itself can be temporally stabilized when the oscillation characteristic itself changes. The temporal fluctuation of the output light intensity may cause deterioration of the transmission signal.
The fourth embodiment of the present invention improves this point.

A fourth embodiment of the present invention will be described in detail with reference to the drawings. FIG. 4 is a block diagram of a fourth embodiment of the direct polarization modulation light source to which the present invention is applied. As shown in the figure, a two-electrode DFB-LD 101, a polarization beam splitter 102, light receiving elements 103-1, 2 and an amplifier 104-1,
2, LPF 105-1, 2, 3, differential amplifier 106-
1, 2, reference voltage source 107, feedback control circuit 108-1,
2. Amplitude control circuit 109, adder 110, drive circuit 11
1-1, 2 and the polarizer 112 as main parts.

The two-electrode DFB-LD 101 is the same as that described in the conventional example, and has the structure shown in FIG. 20 and the characteristics shown in FIG. The difference from the direct polarization modulation light source control method shown in FIG. 1 is that an LPF 105-3, a differential amplifier 106-2,
That is, a reference voltage source 107, a feedback control circuit 108-2, and an amplitude control circuit 109 are added. The amplifier 104-2 was provided with two independent output terminals. The principle of the polarization modulation and the maintenance of the bias point in the switching region (or the maintenance of I _2b + I _2m / 2 near I _2sc in FIG. 22) are the same as in the first embodiment, and will not be described here. In this embodiment, it is assumed that the duty ratio of the modulation signal is fixed at 50%.

Light emitted from the rear end face of the two-electrode DFB-LD 101 is converted into TE light by the polarization beam splitter 102.
It is separated into polarization and TM polarization. The TM polarization is converted into an electric signal by the light receiving element 103-2,
It is amplified by -2. The amplified signal is output to the LPF 105-2 for maintaining the bias point, and is also output to the LPF 105-3 for stabilizing the output. LPF1
05-3 extracts a low frequency component, and outputs a differential amplifier 106-2.
The low frequency component is output to the minus input terminal of. LPF
It is assumed that the cutoff frequency of 105-3 is set sufficiently lower than the modulation signal. The reference voltage source 107 outputs a predetermined voltage signal to a plus input terminal of the differential amplifier 106-2. The differential amplifier 106-2 compares the two signals, and generates a negative voltage when the output of the LPF 105-3 is larger than the output of the reference voltage source 107, and generates a positive voltage when the output is opposite. The feedback control circuit 108-2 generates a control signal using this signal as an error signal, and outputs the control signal to the amplitude control circuit 109. A well-known PID control is preferably used as the feedback control method. Further, the feedback control circuit 108-2 receives the control ON / OFF signal from the transmission unit and turns the output ON / OF.
F. The amplitude control circuit 109 receives the modulation signal from the transmission unit, adjusts the amplitude based on the control signal from the feedback control circuit 108-2, and outputs the adjusted signal to the adder 110. The adder 110 is a control signal from the feedback control circuit 108-1,
The three signals of the modulation signal from the amplitude control circuit 109 and the bias signal from the transmission unit are added and output to the drive circuit 111-1. The drive circuit 111-1 sends a signal to the rear electrode of the two-electrode DFB-LD 101 based on the input from the adder 110.
The drive circuit 111-2 injects a current into each of the front electrodes of the two-electrode DFB-LD 101 based on a bias signal from the transmission unit.

FIG. 5 shows how the output is stabilized in this embodiment. It is assumed that the oscillation characteristics of both polarizations are represented by a dotted curve. If the bias component of the injection current I ₂ is I _2b and the amplitude of the modulation component is I _mod , the polarization modulation is performed between the points A ₀ and B ₀ in FIG. At this time,
At point B ₀ , TE polarization is output at point A ₀ , and TM polarization is output at point P ₀ . Next, it is assumed that the oscillation characteristics of both polarizations have changed to those represented by the solid curve for some reason. In the first embodiment, only the mechanism for maintaining the bias point operates to keep the difference between the output intensities of both polarizations at zero. That is, the bias current component of I ₂ is controlled so that the output intensities of both polarized waves become equal. As a result, the TE polarization operates at point B and the TM polarization operates at point A. However, since the output stability is not guaranteed, the output intensity decreases from P ₀ to P. In this embodiment, the output intensity is kept constant by controlling the modulation amplitude in addition to maintaining the bias point. That is, the amplitude I _{mod of the} modulated signal is controlled so that the output intensity of the TM polarization becomes equal to a certain value.

More specifically, the TM polarization output and the output from the reference voltage source 307-2 are compared by a differential amplifier 306-2, and the obtained signal is processed by a feedback control circuit 308-2. It is input to the control circuit 309. Here, the voltage of the reference voltage source 307-2 is set to a size corresponding to the desired output P _0. The amplitude control circuit 309 generates a modulation signal having an appropriate amplitude based on the control signal from the feedback control circuit 308-2 and the modulation signal from the transmission unit. This modulation signal is added to a bias signal and a control signal for maintaining a bias point by an adder 310, and the driving circuit 311-
After being converted into a current by 1, it is injected into the LD 301. As a result, the modulation amplitude increases from I _mod to I _mod ′, and the output is kept at P ₀ . Thus, the TE polarization is at point B ', TM
Polarization operates at point A ', and polarization modulation is performed between the two points to achieve output stabilization.

[0054]

Embodiment 5 Hereinafter, a fifth embodiment of the present invention will be described with reference to the drawings. FIG. 6 is a block diagram of a fifth embodiment of the direct polarization modulation light source to which the present invention is applied. As shown, a two-electrode DFB-LD 301, a polarization beam splitter 302, light receiving elements 303-1 and 2, an amplifier 304-1,
2, LPFs 305-1 and 2, differential amplifiers 306-1,
2, reference voltage sources 307-1 and 2, feedback control circuit 308-
1 and 2, an amplitude control circuit 309, an adder 310, drive circuits 311-1 and 31-1, and a polarizer 312 as main components.

The two-electrode DFB-LD 301 is the same as the conventional example, and has the structure shown in FIG. 20 and the characteristics shown in FIG. The difference from the fourth embodiment in the configuration is that the output is stabilized for each polarization. Accordingly, reference voltage sources 307-1 and 307-1 serving as outputs are provided for each polarization.

Light emitted from the rear end face of the two-electrode DFB-LD 301 is split into a TE polarization and a TM polarization by the polarization beam splitter 302. The TE polarization is converted into an electric signal by the light receiving element 303-1.
Amplified by The amplified signal is output to the plus input terminal of the differential amplifier 306-1 after the low frequency component is extracted by the LPF 305-1. Reference voltage source 307-1
Outputs a predetermined voltage signal to the minus input terminal of the differential amplifier 306-1. The differential amplifier 306-1 amplifies the difference between the two signals and generates an error signal. The feedback control circuit 308-1 generates a TE control signal based on the error signal and outputs the signal to the adder 310.

The TM polarization split by the polarization beam splitter 302 is processed similarly to the TE polarization. The low-frequency component of the TM polarization output generated by the light receiving element 303-2, the amplifier 304-2, and the LPF 305-2 is output to the minus input terminal of the differential amplifier 306-2. The reference voltage source 307-2 outputs a predetermined voltage signal to the differential amplifier 30.
Output to 6-2 plus input terminal. Differential amplifier 30
6-2 amplifies the difference between the two signals and generates an error signal. The feedback control circuit 308-2 determines TM based on this signal.
A control signal is generated and output to the amplitude control circuit 309.
The feedback control circuits 308-1 and 308-2 output ON / OF signals in response to control ON / OFF signals from the transmission unit.
F. The amplitude control circuit 309 obtains the modulation signal from the transmission unit, adjusts the amplitude based on the TM control signal, and outputs the adjusted signal to the adder 310. The modulated signal after the adjustment is a rectangular wave, and an arbitrary negative voltage corresponds to the mark and OV corresponds to the space (see FIG. 7). The adder 310 is a control signal from the feedback control circuit 308-1, a modulation signal from the amplitude control circuit 309, and a bias signal from the transmission unit.
The two signals are added and output to the drive circuit 311-1.
The drive circuit 311-1 injects a current into the rear electrode of the two-electrode DFB-LD 301 based on the input from the adder 310,
The drive circuit 311-2 injects a current into the front electrode of the two-electrode DFB-LD 301 based on a bias signal from the transmission unit.

FIG. 7 shows how the output is stabilized in this embodiment. It is assumed that the oscillation characteristics of both polarizations are represented by a dotted curve. If the bias component of the injection current I ₂ is I _2b and the amplitude of the modulation component is I _mod ,
I _2b for the modulation mark and I _2b -I for the space
_{A mod} current is injected into the back electrode. At this time, polarization modulation is performed between point A ₀ and point B ₀ in FIG. 7, and TM polarization is output at point A _{0 and} TE polarization is output at intensity P ₀ at point B ₀ . Next, it is assumed that the oscillation characteristics of both polarizations have changed to those represented by the solid curve for some reason. If no feedback is applied to the injection current, the TE polarization operates at point B and the TM polarization operates at point A. At this time, the output intensity of the TE polarization is P
And a difference between the TM polarization and the intensity occurs. In this embodiment,
By controlling the bias component of the injection current, the output of TE polarization is set to P
₀ to keep, maintain the output intensity of the TM polarization to P ₀ and further controls the amplitude of the modulation component. More specifically, the TE polarization output and the output from the reference voltage source 307-1 are compared with the differential amplifier 306-
After comparison with 1, treated with the feedback control circuit 308-1, the bias component of the injection current I ₂ the output intensity of TM polarized waves are controlled so that P _0. In addition, the TM polarization output and the reference voltage source 30
After comparing the output from the 7-2 by the differential amplifier 306-2, a feedback control circuit is treated with 308-2, resulting amplitude of the modulated component of the injection current I ₂ signal based on TE polarization output Control is performed so that the intensity becomes P ₀ . At this time, the bias component of the injection current is I _2b ′, and the amplitude of the modulation component is I _mod ′. As a result, the TE polarization operates at the point B ′ and the TM polarization operates at the point A ′, and polarization modulation is performed between these two points.

[0059]

Embodiment 6 In the fourth and fifth embodiments, the two-electrode DFB-
Although feedback control is performed on the injection current of the rear electrode of the LD, the present invention can be applied to the injection current of the front electrode. Further, in the fourth embodiment, the TM polarization is used to stabilize the output, but the TE polarization may be used. Similarly, in the fourth embodiment, the duty ratio of the modulation signal is 50%.
If not, a reference voltage source and a differential amplifier are newly prepared, and the output of the differential amplifier 106-1 is compared with the output of the reference voltage source by the differential amplifier.
08-1. In the fourth and fifth embodiments, the LPF or the integrator is used as the means for detecting the output intensity of the TE and TM polarizations, but a peak hold circuit may be used as another means.

[0060]

[Embodiment 7] A seventh embodiment of the present invention will be described with reference to FIGS.
The description will be made with reference to FIG. FIG. 8 is a perspective view of the integrated optical semiconductor device, and FIG. 9 is a cross-sectional view along the waveguide. As shown in FIG. 8, a distributed feedback (DFB) laser 1101 having a three-electrode structure is provided in the integrated optical semiconductor device in this embodiment.
A polarization mode splitter 1102 having a Y-branch structure, and a photodetector 1103 (a) for detecting light waves of two different polarization modes separated by the polarization mode splitter 1102.
1103 (b) are integrated. These waveguides are
It is formed with a buried structure 1106 of high resistance InP. In this embodiment, the polarization mode splitter 1102
In this method, the superlattice structure forming the waveguide is mixed-crystallized by loading and annealing the SiN film only in the region 1104 (a), so that only the TM mode light wave is guided. I have. Therefore, the lightwave emitted from the DFB laser 1101 is converted into a TE mode lightwave by the polarization mode splitter 1102 into the waveguide 1104 (a).
The M-mode light wave is split into a waveguide 1104 (b),
Each is detected by the photodetectors 1103 (a) and 1103 (b) and converted into an electric signal.

Next, an element structure and a manufacturing method of this embodiment will be described with reference to FIG. In FIG. 9, reference numeral 1201 denotes an n-InP layer serving as a substrate; and 1202, a depth of 0.05 μm.
N-InP lower cladding layer having a diffraction grating formed thereon, 1
203 is a 0.2 μm n-In _0.71 Ga _0.29 As _0.62 P
_0.38 Lower guide layer, 1204 is well layer i-In _0.53 Ga
_0.47 As (5 nm thick), barrier layer i-In _0.59 Ga
An active layer having a strained superlattice structure consisting of 10 layers of _0.41 As (5 nm in thickness), 1205 is a p-InP cladding layer, 1205 is
6 is a p-In _0.59 Ga _0.41 As _0.9 P _0.1 contact layer,
Reference numerals 1208 and 1208 'denote electrode separation regions from which the contact layer 1206 has been removed, 1209 and 1209' denote Cr / AuZnNi / Au layers which are p-type electrodes, and 1210: Cr / AuZnNi / which are electrodes through which a current in which a signal is superimposed flows. A
u layer 1211 is AuGeNi / Au which is a substrate side electrode
Layers 1220 are AuGeNi / A electrodes for photodetectors.
u layer, 1212 and 1212 ′ are Si to be an anti-reflection film
The O ₂ film is shown.

Part of DFB laser 1101 and Y-branch conductor
In the waveguide portion 1102, the contact layer, the cladding
After etching the layer and the active layer, i-In_0.53Ga_0.47
As well layer (thickness 3 nm), i-InP barrier layer (thickness
5 nm) Light guide layer 121 having a superlattice structure composed of 20 layers
3, p-InP cladding layer 1214, p-In_0.59Ga
_0.41As_0.9P_0.1Contact layer 1215 for selective regrowth
Is formed. Regrown light guide layer 1213
Has a wavelength corresponding to an energy band gap of about 1.
3 μm, for light with a laser oscillation wavelength of 1.55 μm.
In this case, the structure is small.

In this DFB laser, the active layer 12
Since 04 is in multiple quantum well layer having a tensile _strain, E lh0 -E _e0 transition has energy almost equal in E _hh0 -E _e0, compared to conventional DFB laser TM
The configuration is such that the oscillation threshold value in polarization is low and polarization switching can be performed efficiently. In this manner, the light wave modulated by the polarization switching is emitted from the front end face 1105 of the element, becomes a signal ASK-modulated by extracting only the TE mode by the polarizer 1112, enters the optical fiber 1113, and is transmitted. You.

Here, the pitch of the diffraction grating is set to 240 nm so that the distributed feedback wavelength by the diffraction grating of the DFB laser unit is _close to the wavelength corresponding to E _lh0 -E _e0 , 1.562 μm in the TE mode, and TM mode. And a configuration having a Bragg wavelength of 1.558 μm.

The DFB laser unit 1101 has other structures, for example, a structure without a diffraction grating in a region without an active layer, a structure in which a diffraction grating has a phase shift, and a structure with an active layer in all regions. And so on. For simplicity, the above-mentioned two-electrode DFB laser may be used.

Next, a method of driving the present element will be described with reference to FIGS. This embodiment adopts the light source control method of the first embodiment. In FIG. 1, the DFB laser 1-1, the light receiving elements 1-3-1, 1-3-2, etc. are three-electrode DFB-LD.
1101, detectors 1103 (a), 1103 (b) and the like.

When the DFB laser 1101 is driven, the DF
Light waves are emitted from the front and rear two end faces from the B laser 1101. Outgoing light from the rear end face is extracted only by TE polarization by the polarizer 1112.

The DFB laser 11 is emitted from the front end face.
01 and the polarization beam splitter 1102 integrated with T
The light waves separated into the E mode and the TM mode are converted into electric signals by the integrated photodetectors 1103 (a) and 1103 (b) and amplified by the amplifiers 1-4-1 and 1-4-2. Then, only the low frequency part is extracted by the LPFs 1-5-1 and 2. These and the following are substantially the same as those described in the first embodiment. Finally, the driving circuit 1-10
-1 outputs a current corresponding to the input signal, and this current is D
It is input to the electrode 1210 of the FB laser 1101. The drive circuit 1-10-2 outputs a current corresponding to the bias signal input from the transmission unit, and the current is output to the DFB laser 11
01 electrodes 1209 and 1209 ′. The operation of the driving method is also substantially the same as that of the first embodiment, and a description thereof will be omitted.

In this embodiment, the polarization beam splitter 110
The method of loading and annealing the SiN film was described as the manufacturing method of the second method.
The same thing can be obtained even if a mixed crystal is formed by diffusion. Alternatively, a polarization mode splitter can be obtained by making the Y branch an asymmetric Y branch and loading a metal thin film on one of the waveguides. In this configuration, the waveguide loaded with the metal thin film has the TE mode, and the other waveguide has the TM mode.
The mode propagates.

In this embodiment, feedback control is performed on the current injected into the same electrode (electrode 1210) as the electrode through which the modulation signal flows.
09 ′) can also be applied.

[0071]

Embodiment 8 An eighth embodiment according to the present invention will be described with reference to FIG. FIG. 10 shows that the LiNbO ₃ substrate 1401 has Y
A branch-type polarization mode splitter 1404 is formed, and a three-electrode DFB laser 1402 and a photodiode 140
3 (a) and 1403 (b) are attached to a substrate 1404 with an ultraviolet curable resin or the like and mounted on a hybrid.
The polarization mode splitter 1404 loads a metal thin film 405 on one of the Y-branch waveguides, and the TE mode propagates to this side. The TM mode propagates to the other waveguide, and each output is a photodiode 140
3 (a) and 1403 (b). DFB
The laser 1402 has the same structure as that of the seventh embodiment, and the light wave emitted from the rear end face is directed to the polarization mode splitter 1404.
The light wave emitted from the front end face is used as an optical signal. Hybrid mounting optical coupling section and DFB laser 14
02 has a non-reflection coating on the front emission end to suppress reflection. The driving method is the same as in the seventh embodiment.

[0072]

Ninth Embodiment A ninth embodiment according to the present invention is shown in FIG. The present invention utilizes a vertical type forward coupler as a polarization mode splitter and is monolithically integrated on a semiconductor substrate.

In FIG. 11, 1501 is an n-InP substrate, 1502 is an n-InP buffer layer, 1503 is an n-InGaAsP lower waveguide layer having a band gap wavelength of 1.1 μm, 1504 is an n-InP cladding layer, and 1505 is a band. An n-InGaAsP light guide layer having a gap wavelength of 1.1 μm, and a pitch 1506 serving as a DFB laser unit
nm, 1507 is a diffraction grating with a pitch of 14.5 μm to be a forward coupler, 1508 is an n-InGaAsP upper waveguide layer with a band gap wavelength of 1.3 μm, 150
9 is a well layer i-In _0.53 Ga _0.47 As (thickness: 5 nm);
Barrier layer i-In _0.28 Ga _0.72 As (thickness 5 nm) 10
An active layer having a strained superlattice structure composed of layers, 1510 is p-In
P cladding layer, 1511 has a band gap wavelength of 1.5μ
m, p-InP contact layer, 1512, p-InP cladding layer, 1513, p-InP contact layer with a band gap wavelength of 1.5 μm, 1514, p-type electrode, 151
5 is an n-type electrode, 1516 and 1517 are antireflection coatings, and 1518 is a photodetector. The photodetector 1518 may be a surface-type photodetector that is attached on the antireflection coating 1517.

Next, the operation of the present apparatus will be described. D
The FB laser section is polarization-modulated similarly to the seventh embodiment. In the forward coupler section, the light A propagating through the upper waveguide layer 1508
Is coupled (B) to the lower waveguide layer 1503 and propagated. FIG. 12 shows a coupling characteristic of the forward coupler from the upper waveguide layer 1508 to the lower waveguide layer 1503. As shown in this figure, for the TE mode light, the laser oscillation wavelength is 1.5.
The filter characteristic has a center wavelength of 5 μm. Its half width is 5 nm. On the other hand, a light wave of the TM mode has a characteristic having a center wavelength on a shorter wavelength side by about 30 nm than this. Therefore, the laser oscillation wavelength of 1.55
The light of μm propagates through the lower waveguide layer 1503 only in the TE mode, and is received by a photodetector 1518 provided outside. On the other hand, the TM light propagates through the upper waveguide layer 1508 as it is, and is detected by the light detection unit in the substrate 1501. At this time, in order to cut the TM light leaking from the rear part of the light detector, the end face of the rear part of the light detector may be damaged by etching.

The driving method of the device is the same as that of the seventh embodiment. The lightwave emitted from the front end face of the DFB laser is used as a signal, the TE mode component of the lightwave emitted from the rear end face is detected by the light detection section, and the TM mode component is detected by the photodetector 1.
After being detected at 518 and converted into an electric signal, it is used for feedback control.

In this forward coupler, the center wavelength of filtering can be changed by applying an electric field. Therefore, if the center wavelength is shifted in accordance with the change in the oscillation wavelength of the laser, the coupling efficiency can be made constant.

[0077]

Embodiment 10 FIG. 13 shows a tenth embodiment of the present invention. This embodiment is an example in which a directional coupler is used as a polarization mode splitter.

In FIG. 13, reference numeral 1601 denotes a seventh embodiment of the present invention.
A DFB laser having the same structure as that of the embodiment, 1602 is a directional coupler having a coupling length of 1 (ell), 1603 (a) and 1603
3 (b) is a photodetector. The directional coupler 1602 is
Waveguide 1604 having a shape as shown by etching
(A) After forming 1604 (b), a high-resistance InP
It is formed by embedding the layer 1605.

The polarization-modulated light wave emitted from the DFB laser 1601 enters the directional coupler 1602.
Waveguide 1604 constituting directional coupler 1602
Since the equivalent refractive index in (a) and 1604 (b) differs between the TE mode light wave and the TM mode light wave emitted from the DFB laser 1601, the coupling efficiency in the directional coupler is T.
It differs between E mode and TM mode. Therefore, by setting the coupling length 1 (ell) of the directional coupler such that the light wave of the TE mode is coupled to the waveguide 1604 (a) and the light wave of the TM mode is coupled to the waveguide 1604 (b), It can be used as the mode splitter 1602. The light waves separated in this manner are respectively converted into light detectors 1603
(A) and 1603 (b) are converted into electric signals. The driving method can be performed in the same manner as in the seventh embodiment.

[0080]

Embodiment 11 FIG. 14 shows an eleventh embodiment of the present invention. This embodiment is an example in which an etching mirror provided in an integrated optical device is used as a polarization mode splitter.

In FIG. 14, reference numeral 1701 denotes a seventh embodiment of the present invention.
DFB laser having the same structure as the embodiment, 1702 is DF
Angle θ _{b with} respect to the light wave emitted from B laser 1701
1703, an etching mirror configured to be
(A) and 1703 (b) are photodetectors. The etching mirror 1702 is formed by forming a fine etching groove by reactive ion beam etching. In order to adjust the reflectance of the reflector, a substance having an appropriate refractive index may be inserted into the etching groove.

Next, the operation of this embodiment will be described.
Polarization modulated light waves are emitted from the DFB laser 1701 is incident on the etched mirror 1702 at an incident angle theta _b. Here, if the incident angle θ _b is set to be a Brewster angle with respect to the oscillation light of the DFB laser 1701, the TE mode light wave is transmitted without reflection loss,
Most of the TM mode light wave is reflected by the etching mirror, and functions as a polarization mode splitter. The light waves separated in this way are respectively detected by photodetectors 1703.
(A) and 1703 (b) convert the signal into an electric signal. The driving method can be performed in the same manner as in the seventh embodiment.

[0083]

Embodiment 12 FIG. 15 shows a twelfth embodiment of the present invention. This embodiment is an example in which the polarization mode splitter shown in the seventh embodiment of the present invention is integrated at two places on the front side and the rear side of the DFB laser.

In FIG. 15, reference numeral 1801 denotes a DFB laser which is the same as that of the seventh embodiment; 1802 and 1802 ', a polarization mode splitter using the same Y-branch as that of the seventh embodiment;
(A) and 1803 (b) are photodetectors.

In the integrated optical semiconductor device shown in the seventh to eleventh embodiments of the present invention, the polarization-modulated signal light is
ASK modulation is performed by emitting light from the front end face of the laser and extracting only TE (or TM) mode light waves by a polarizer provided at the emission end. In such a configuration, it is difficult to reduce the size of the module, and optical adjustment is required at the time of modularization, resulting in poor productivity. Therefore, as in the present embodiment, the polarization mode splitter 1 is disposed on the end face side for outputting the modulation signal in the same manner as the opposite side.
By integrating 802 ', it is possible to reduce the size of the module and improve the productivity.

As described above, for the two polarization mode splitters 1802 and 1802 'provided in the integrated optical semiconductor device, all of the eighth to eleventh embodiments of the present invention can be applied. Needless to say, a combination of these is also possible. In addition, a semiconductor optical amplifier may be integrated at the emission end in order to compensate for the power of a light wave extracted outside as a light signal from the polarization mode splitter.

[0087]

Embodiment 13 In the seventh to twelfth embodiments, the integrated optical semiconductor device is controlled by the driving method of the first embodiment. However, the integrated optical semiconductor device of the thirteenth embodiment of the present invention is different from that of the second embodiment. The driving method (FIG. 3) is used. The integrated optical semiconductor device of any one of the seventh to twelfth embodiments may be controlled by the driving method of the second embodiment.

[0088]

Embodiment 14 FIG. 16 shows a fourteenth embodiment of the present invention. FIG. 16A is a diagram for explaining a basic concept, and FIG. 16B is a diagram showing a configuration of an integrated device configured according to the concept of FIG. 16A.

In FIG. 16A, reference numeral 2001 denotes a TE / TM polarization modulation laser capable of changing the polarization state of output light depending on the state of excitation, 2002 a semiconductor laser in which TE light is output light, and 2003 a TM laser. A semiconductor laser whose light is output light; 2004 and 2005 are photodetectors;
6 is an optical branching and joining element, 2007 is a polarizer, and 2008 is T
Output light of E / TM polarization modulation laser, 2009 is output light 2
Of the two orthogonal polarizations 008, an optical signal transmitted by only one of them by the polarizer 2007, 2010 is an optical fiber, and 2011 is a control circuit.

The TE / TM polarization modulation laser 2001 is a DFB semiconductor laser having the same configuration as that of the seventh embodiment. The TE oscillation laser 2002 is a DFB laser (or DBR laser) that outputs TE-polarized light at the same wavelength as the TE / TM polarization modulation laser 2001 at the time of TE oscillation. The TM oscillation laser 2003 outputs a TM-polarized light at the same wavelength as that of the TE / TM polarization modulation laser 2001 at the time of TM oscillation.
BR laser). The TE oscillation laser 2002 and the TM oscillation laser 2003 can be simply configured to have the same configuration as the TE / TM polarization modulation laser 2001 and to satisfy the above conditions by setting the operating point. Optical branching and joining element 2
006 is the output light of the TE oscillation laser 2002 and TE / T
The output light of the M polarization modulation laser 2001 is detected by a photodetector 2004.
At the same time, and the beat signal is obtained. Further, the output light of the TM oscillation laser 2003 and the output light of the TE / TM polarization modulation laser 2001 are simultaneously detected by the photodetector 2005, and their beat signals can be obtained.

Next, the operation will be described with reference to FIG. The TE / TM polarization modulation laser 2001 includes a control circuit 2
011 is polarization-modulated. Further, the TE oscillation laser 2002 and the TM oscillation laser 2003 are driven by the control circuit 2011, and are each driven so as to obtain the above-described operation. In this operating state, the photodetector 20
In 04, a beat signal of TE polarized light is obtained, and in the photodetector 2005, a beat signal of TM polarized light is obtained. These function as the polarization mode splitter and the polarization beam splitter. These two photodetectors 2004,
Based on the output signal of 2005, the control circuit 2011 has, for example, a configuration as shown in the first embodiment (FIG. 1), the second embodiment (FIG. 3), the fourth embodiment (FIG. 4), and the like. And TE /
By controlling the TM polarization modulation laser 2001, a stable operation can be obtained. However, unlike the control system of these embodiments,
In the control circuit of the present embodiment, the two photodetectors 2004, 2
It also has a function of stabilizing the operation of the TE oscillation laser 2002 and the TM oscillation laser 2003 so that a beat signal is obtained in 005.

FIG. 16 (b) showing the above configuration as an integrated device, and the same parts as those in FIG. 16 (a) are denoted by the same reference numerals. The configuration includes three embedded DFB lasers (2002, 2002, 2003), an embedded photodetector (2004, 2005), and a branching / joining waveguide 2006 connecting these. For the DFB laser portion and the photodetector portion, for example, the same configuration as in the seventh embodiment is used. As the branching / combining waveguide 2006, the waveguide of the seventh embodiment in which the superlattice structure is not mixed-crystallized is used (that is, one having no polarization dependence). By configuring such an integrated device, an element that exhibits the operation described with reference to FIG.

[0093]

Embodiment 15 In a multi-electrode DFB laser used for polarization modulation, the wavelength also switches with the switching between TE polarization and TM polarization by modulation (the wavelength difference between TE polarization and TM polarization is up to 1 nm). In the first embodiment and the like, the polarization at the time of modulation is separated and the modulation state is appropriately maintained based on the difference in the intensity. However, it is also possible to separate two polarizations according to the wavelength. That is, the part such as the polarization beam splitter is
A splitter 401 (divides input light into two outputs at the center of the center of the wavelength of the TE polarization and the wavelength of the TM polarization) shown in FIG.
Alternatively, it is replaced with an optical splitter 402 and two optical filters 4031-1 and 4032-1 shown in FIG. Duplexer 40
1 and the characteristics of the optical filters 403-1 and 403-2 are shown in FIG.
7 (c) and (d).

[0094]

Embodiment 16 FIG. 18 is a block diagram of a wavelength division multiplexing optical communication system according to a sixteenth embodiment of the present invention. This system is preferably used for video distribution. The number of channels is n, the number of receiving devices 505 is m, and the wavelength of each channel is λ ₁ , λ ₂ ,
Shown in ~λ _n. The transmitting device 501 modulates each of the n light sources with a digital video signal and transmits the modulated light to the optical fiber 502. This optical signal is sent to the optical splitter 503 by m optical fibers 5.
04-1 to 04-m and sent to the receiving devices 505-1 to 505-1. The transmission device 501 includes a plurality of direct polarization modulation light sources or integrated semiconductor devices to which the control method of the above-described embodiment is applied.

By using the direct polarization modulation light source control method of the present invention, a wavelength multiplexing optical communication system in which the occupied optical frequency of one channel is small and the modulation state is good can be realized by the simple method of direct modulation.

[0096]

According to the structure of the present invention, in the direct polarization modulation light source control method, it is possible to provide a control method in which the modulation state is kept good, the occupied light frequency is small and the structure is simple. Further, the output light intensity can be kept constant.

Further, in the method for driving an integrated optical semiconductor device according to the present invention, a compact and high-productivity optical communication system for a high-density wavelength division multiplexing optical communication system using a direct steep wave modulation method having a very small dynamic wavelength variation. A method for driving a light source for optical communication can be provided.

[Brief description of the drawings]

FIG. 1 is a block diagram of a light source control method according to a first embodiment of the present invention.

FIG. 2 is a schematic diagram showing a relationship between I _2b and an error signal in the first embodiment.

FIG. 3 is a block diagram of a light source control method according to a second embodiment of the present invention.

FIG. 4 is a block diagram of a light source control method according to a fourth embodiment of the present invention.

FIG. 5 is a schematic diagram for explaining output stabilization in a fourth embodiment.

FIG. 6 is a block diagram of a light source control method according to a fifth embodiment of the present invention.

FIG. 7 is a schematic diagram for explaining output stabilization in a fifth embodiment.

FIG. 8 is a perspective view of an integrated semiconductor device according to a seventh embodiment of the present invention.

FIG. 9 is a sectional view of a seventh embodiment.

FIG. 10 is a perspective view of an integrated semiconductor device according to an eighth embodiment of the present invention.

FIG. 11 is a sectional view of an integrated semiconductor device according to a ninth embodiment of the present invention.

FIG. 12 is a diagram illustrating characteristics of a polarization mode splitter according to a ninth embodiment.

FIG. 13 is a perspective view of an integrated semiconductor device according to a tenth embodiment of the present invention.

FIG. 14 is a perspective view of an integrated semiconductor device according to an eleventh embodiment of the present invention.

FIG. 15 is a perspective view of an integrated semiconductor device according to a twelfth embodiment of the present invention.

FIG. 16 is a diagram showing a basic conceptual configuration and a specific structure of an integrated semiconductor device according to a fourteenth embodiment of the present invention.

FIG. 17 is a diagram showing characteristics of a main part, a duplexer, and an optical filter according to a fifteenth embodiment of the present invention.

FIG. 18 is a configuration diagram of an optical communication system according to a sixteenth embodiment of the present invention.

FIG. 19 is a block diagram of a proposed example of a direct polarization modulation light source.

FIG. 20 is a sectional perspective view of a two-electrode DFB-LD for direct polarization modulation.

FIG. 21 is a schematic diagram of the oscillation characteristics of a two-electrode DFB-LD for direct polarization modulation.

FIG. 22 is a schematic diagram showing a state of switching between TE polarization and TM polarization.

FIG. 23 is a diagram showing an optical output when a switching region is shifted.

FIG. 24 is a schematic diagram showing a state of switching between TE polarization and TM polarization.

FIG. 25 is a schematic diagram showing a change in optical output accompanying switching between TE polarization and TM polarization.

FIG. 26 is a schematic diagram for explaining a mechanism for maintaining a bias point.

FIG. 27 is a schematic diagram for explaining a change in optical output accompanying a change in oscillation characteristics.

[Explanation of symbols]

1-1, 101, 301 2-electrode DFB-LD 1-2, 102, 302 Polarization beam splitter 1-3, 103, 303 Photodetector 1-4, 104, 304 Amplifier 1-5, 105, 305 LPF 1 6, 106, 306 Differential amplifier 1-7, 108, 308 Feedback control circuit 1-8 Switch 1-9, 110, 310 Adder 1-10, 111, 311 Drive circuit 1-11, 112, 312 Polarizer 3 -1 Subtractor 3-2, 107, 307 Reference voltage source 109, 309 Amplitude control circuit 401 Demultiplexer 402 Splitter 403 Optical filter 501 Transmitter 502, 504, 1113, 2010 Optical fiber 503 Optical splitter 505 Receiver 1101 , 1402, 1601, 1701, 1801,
2001, 2002, 2003 Semiconductor DFB lasers 1102, 1404, 1602, 1802, 1802 '
Polarization mode splitters 1103, 1403, 1518, 1603, 1703
1803, 2004, 2005 Photodiode 1104, 1604 Waveguide 1105 Outgoing end face 1106, 1605 Buried layer 1112, 2007 Polarizer 1201, 1401, 1501 Substrate 1202, 1205, 1214, 1504, 1510,
1512 Cladding layer 1203, 1213, 1505 Guide layer 1204, 1509 Active layer 1206, 1215, 1511, 1513 Contact layer 1208, 1208 'Electrode separation region 1209, 1209', 1210, 1211, 151
4,1515 Electrode 1212,1212 ', 1516,1517 Antireflection film 1405 Metal thin film 1502 Buffer layer 1503,1508 Waveguide layer 1506 Fine pitch diffraction grating 1507 Coarse pitch diffraction grating 1702 Etching mirror 2006 Optical merging / branching element 2011 Control circuit

──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Toshihiko Ouchi 3-30-2 Shimomaruko, Ota-ku, Tokyo Inside Canon Inc. (72) Inventor Atsushi Nitta 3-30-2 Shimomaruko, Ota-ku, Tokyo (56) References JP-A-62-242593 (JP, A) JP-A-4-355986 (JP, A) JP-A-4-113327 (JP, A) JP-A-4-73606 ( JP, A) (58) Field surveyed (Int. Cl. ⁷ , DB name) H01S 5/00-5/50

Claims (9)

(57) [Claims] 1. By directly modulating the amount to be driven,
A control method of a direct polarization modulation light source capable of modulating output light into two polarizations whose polarization planes are orthogonal to each other, comprising: means for separating the two orthogonal polarizations of output light of the direct polarization modulation light source;
Means for receiving the two orthogonally polarized waves separated from each other,
Means for differentially amplifying two signals from the light receiving means,
A direct polarization modulation light source control method, wherein feedback control is performed so that the intensity difference between the two orthogonal polarizations becomes a predetermined value. 2. By directly modulating the amount to be driven,
A control method of a direct polarization modulation light source capable of modulating output light into two polarizations whose polarization planes are orthogonal to each other, comprising: means for separating the two orthogonal polarizations of output light of the direct polarization modulation light source;
Means for receiving the two orthogonally polarized waves separated from each other,
Means for taking the difference between the two signals from the means for receiving light, a reference voltage source, means for differentially amplifying the output of the means for taking the difference and the output of the reference voltage source, and A direct polarization modulation light source control method, wherein feedback control is performed so that the intensity difference becomes constant. 3. By directly modulating the amount to be driven,
A control method of a direct polarization modulation light source capable of modulating output light into two polarizations whose polarization planes are orthogonal to each other, comprising: means for separating the two orthogonal polarizations of output light of the direct polarization modulation light source;
Means for receiving the two orthogonally polarized waves separated from each other,
Differential amplification of two signals obtained from the light receiving means, a reference voltage source, and a signal obtained from one of the polarized light receiving means and a signal from the reference voltage source. Means for performing feedback control so that the intensity difference between the two orthogonal polarizations becomes a predetermined value.
A direct polarization modulation light source control method, wherein feedback control is performed so that the output intensity of one of the two polarized waves is constant. 4. By directly modulating the amount to be driven,
A control method of a direct polarization modulation light source capable of modulating output light into two polarizations whose polarization planes are orthogonal to each other, comprising: means for separating the two orthogonal polarizations of output light of the direct polarization modulation light source;
Means for receiving the two orthogonally polarized waves separated from each other,
A reference voltage source for each of the two orthogonal polarizations, and a unit for differentially amplifying a signal obtained from a unit for receiving the polarization and a signal from the reference voltage source, respectively, A direct polarization modulation light source control method, wherein feedback control is performed so that the output intensities of the waves become constant. 5. The direct polarization modulation light source control method according to claim 1, wherein the predetermined value is zero. 6. A multi-electrode DFB as a direct polarization modulation light source
(Distributed Feedback) -LD
The direct polarization modulation light source control method according to any one of claims 1 to 5, wherein (Laser Diode) is used. 7. The direct polarization modulation light source control method according to claim 1, wherein the means for separating the two orthogonal polarizations is constituted by a polarization beam splitter. 8. The apparatus according to claim 1, wherein said means for separating two orthogonal polarizations comprises means for separating two orthogonal polarizations by wavelength. Direct polarization modulation light source control method. 9. A semiconductor laser having a structure in which two orthogonal polarization modes are switched by modulating a current flowing through a part of an optical waveguide including a light emitting layer, and the orthogonal of modulated output light of the semiconductor laser. Integrated optical semiconductor device in which means for electrically detecting the output light intensities of the two polarization modes are integrated on the same substrate, in order to stabilize the modulation state of the light wave emitted from the semiconductor device to the outside. Means for taking the difference between the electric signals obtained by converting the output light intensities of the two polarization modes,
A reference voltage source, a means for differentially amplifying the output of the means for taking the difference and the output of the reference voltage source, and providing the semiconductor laser with a constant intensity difference between the two polarized waves having different polarization planes. A method for driving an integrated optical semiconductor device, wherein feedback control is performed.