JP4365416B2 - Optical delay detection circuit for optical multilevel modulation signal - Google Patents

Description

  The present invention relates to an optical delay detection circuit that delay-detects a phase-modulated optical signal, and more particularly to a technique for realizing an inexpensive optical delay detection circuit for an optical multilevel modulation signal.

  In recent years, there has been an increasing demand for an increase in transmission capacity in an optical transmission system such as an increase in Internet traffic and a fusion of broadcasting and communication. When the optical multi-level modulation method is used, good characteristics such as narrow spectral width, high dispersion tolerance, and relaxation of electric circuit operation speed are expected. Therefore, studies on transmission systems have been actively conducted. In particular, there are many reports that use an optical circuit with a built-in optical delay circuit that does not use local light for the construction of an inexpensive and practical transmission system. In addition, the differential reception method using the photoelectric element pair has a feature that it is excellent in consistency with a next-generation optical network in which the optical path is switched because it is not necessary to control the threshold level even if the received light intensity changes. For this reason, research is actively being conducted on an optical delay detection circuit that has a built-in optical delay circuit and performs differential reception with a pair of photoelectric elements.

(First prior art example)
As an optical delay detection circuit for detecting an optical multilevel phase modulation signal, for example, a circuit shown in FIG. 26 is known (see, for example, Non-Patent Document 1).

  In the circuit shown in FIG. 26, optical two branch circuits 1801a and 1801b are connected in parallel to the optical N branch circuit 1805. The optical branching circuit 1801a is connected to delay time difference fine adjustment function units 1802a and 1802b. Here, the delay time difference fine adjustment function units 1802a and 1802b are collectively referred to as a first optical delay unit 1803a. The outputs of the delay time difference fine adjustment function units 1802a and 1802b are connected to a 2-input 2-output optical multiplexing / demultiplexing circuit 1806a. The two outputs of the 2-input 2-output optical multiplexing / demultiplexing circuit 1806a are connected to the photoelectric conversion element pair 1804a.

  Similarly, delay time difference fine adjustment function units 1802c and 1802d are connected to the optical branching circuit 1801b. Here, the delay time difference fine adjustment function units 1802c and 1802d are collectively referred to as a first optical delay unit 1803b. The outputs of the delay time difference fine adjustment function units 1802c and 1802d are connected to the 2-input 2-output optical multiplexing / demultiplexing circuit 1806b. The two outputs of the 2-input 2-output optical multiplexing / demultiplexing circuit 1806b are connected to the photoelectric conversion element pair 1804b.

  In the circuit according to the first prior art example, a plurality of Mach-Zehnder optical interference circuits are arranged in parallel, and the delay amounts of the Mach-Zehnder optical interference circuits are different from each other in the range of 0 to λ / c. Is set to a value. Then, after receiving the optical signal that has passed through each Mach-Zehnder optical interference circuit by the photoelectric element, the electrical signal generated by the optical signal is logically operated as necessary, and the optical multilevel signal is converted into a multi-channel electrical signal. .

(Second prior art example)
Further, as shown in FIG. 27, for example, an optical multilevel phase modulation signal is detected by attaching a 2-input 4-output MMI (Multi-Mode Interference) waveguide to the end of the optical delay circuit. An optical delay detection circuit has been proposed (see, for example, Patent Document 1).

  In the circuit shown in FIG. 27, the delay time difference fine adjustment function units 1902a and 1902b are connected to the optical branching circuit 1901. Here, the delay time difference fine adjustment function units 1902a and 1902b are collectively referred to as a first optical delay unit 1903. The outputs of the delay time difference fine adjustment function units 1902a and 1902b are connected to a 2-input 4-output MMI optical multiplexing / demultiplexing circuit 1919. Of the four outputs of the 2-input 4-output MMI optical multiplexing / demultiplexing circuit 1919, two outputs are connected to the photoelectric conversion element pair 1904a, and the remaining two outputs are connected to the photoelectric conversion element pair 1904b. Has been.

  In the second prior art example, only one interference circuit performs the role played by the plurality of Mach-Zehnder optical interference circuits in the first prior art example.

(Third prior art example)
Further, for example, an optical delay detection circuit shown in FIG. 28 is proposed as a device for detecting an optical multilevel phase modulation signal by attaching a 2-input 4-output star coupler type optical multiplexing / demultiplexing circuit to the end of the optical delay circuit. (For example, see Non-Patent Document 2).

  In the circuit shown in FIG. 28, the delay time difference fine adjustment function units 2002a and 2002b are connected to the optical branching circuit 2001. Here, the delay time difference fine adjustment function units 2002a and 2002b are collectively referred to as a first optical delay unit 2003. The outputs of the delay time difference fine adjustment function units 2002a and 2002b are connected to a 2-input / 4-output star coupler type optical multiplexing / demultiplexing circuit 2018. Of the four outputs of the star coupler type optical multiplexing / demultiplexing circuit 2018, two outputs are connected to the photoelectric conversion element pair 2004a, and the remaining two outputs are connected to the photoelectric conversion element pair 2004b. .

  In the third prior art example, as in the second prior art example, only one interference circuit performs the role played by the plurality of Mach-Zehnder optical interference circuits in the first prior art example.

(Fourth prior art example)
On the other hand, for example, an optical delay detection circuit as shown in FIG. 29 is studied as an optical circuit for delay detection of a multi-level modulation signal whose polarization state is modulated (see, for example, Non-Patent Document 3). .

  In the circuit shown in FIG. 29, optical branching circuits 2101 a and 2101 b are connected in parallel to the optical N branching circuit 2105. A fast axis 45-degree inclined half-wave plate 2133 and an optical path are connected to the optical branching circuit 2101a. Here, the fast axis 45-degree inclined half-wave plate 2133 and the optical path are collectively referred to as a first optical delay unit 2103a. The fast axis 45 degree inclined half-wave plate 2133 and the output of the optical path are connected to a 2-input 2-output optical multiplexing / demultiplexing circuit 2106a. The two outputs of the 2-input 2-output optical multiplexing / demultiplexing circuit 2106a are connected to the photoelectric conversion element pair 2104a.

  Similarly, a fast axis −45 ° inclined half-wave plate 2134 and an optical path are connected to the optical branching circuit 2101b. Here, the fast axis-45 degree inclined half-wave plate 2134 and the optical path are collectively referred to as a first optical delay unit 2103b. The fast axis-45 degree inclined half-wave plate 2134 and the output of the optical path are connected to a 2-input 2-output optical multiplexing / demultiplexing circuit 2106b. The two outputs of the 2-input 2-output optical multiplexing / demultiplexing circuit 2106b are connected to the photoelectric conversion element pair 2104b.

(Fifth prior art example)
There is also known an optical delay detection circuit that delay-detects an intensity-modulated differential optical signal. For example, it has been studied to omit a circuit for controlling the threshold level using a circuit as shown in FIG. 30 (see, for example, Non-Patent Document 4 and Non-Patent Document 5).

  In the circuit shown in FIG. 30, two optical paths are connected to the optical branching circuit 2201. Here, the two optical paths are collectively referred to as a first optical delay unit 2203. The outputs of these two optical paths are directly connected to the photoelectric conversion element pair 2204.

(Sixth prior art example)
On the other hand, for example, the circuit shown in FIG. 31 has been reported as one that matches the frequency characteristics of the Mach-Zehnder optical interference circuit with the wavelength of the input signal light (see, for example, Non-Patent Document 6).

In the circuit shown in FIG. 31, delay time difference fine adjustment function units 2302 a and 2302 b are connected to the optical branching circuit 2301. Here, the delay time difference fine adjustment function units 2302a and 2302b are collectively referred to as a first optical delay unit 2303. The outputs of the delay time difference fine adjustment function units 2302 a and 2302 b are connected to a 2-input 2-output optical multiplexing / demultiplexing circuit 2306. The two outputs of the 2-input 2-output optical multiplexing / demultiplexing circuit 2306 are connected to the photoelectric conversion element pair 2304. Further, the output of the photoelectric conversion element pair 2304 is connected to the low-pass filter 2320, the output of the low-pass filter 2320 is connected to the integrator 2310, and the output of the integrator 2310 is a delay time difference slightly smaller. It is connected to the adjustment function unit 2302a.
This circuit has a feature that it operates even with a very simple control circuit.

(Seventh prior art example)
Further, for example, an optical delay detection circuit shown in FIG. 32 has been reported as one in which the frequency characteristic itself of the Mach-Zehnder optical interference circuit is minutely oscillated to synchronize with the wavelength of the input light (for example, patent literature). 2).

  In the circuit shown in FIG. 32, the delay time difference fine adjustment function units 2402a and 2402b are connected to the optical branching circuit 2401. Here, the delay time difference fine adjustment function units 2402a and 2402b are collectively referred to as a first optical delay unit 2403. The outputs of the delay time difference fine adjustment function units 2402a and 2402b are connected to a 2-input 2-output optical multiplexing / demultiplexing circuit 2406. The two outputs of the 2-input 2-output optical multiplexing / demultiplexing circuit 2406 are connected to the photoelectric conversion element pair 2404. The output of the photoelectric conversion element pair 2404 is connected to the mixer 2423, the output of the mixer 2423 is connected to the low-pass filter 2420, and the output of the low-pass filter 2420 is connected to the adder 2422. The output of the adder 2422 and the output of the integrator 2410 are connected to the delay time difference fine adjustment function unit 2402a.

  One of the two outputs of the low frequency oscillator 2424 is connected to the adder 2422, and the other one of the two outputs of the low frequency oscillator 2424 is connected to the phase adjustment circuit 2421. The output of the phase adjustment circuit 2421 is connected to the mixer 2423.

  This circuit is widely used to adjust the center wavelength of a WDM (Wavelength Division Multiplexing) filter to the input signal light wavelength.

(Eighth prior art example)
Furthermore, although it was studied for an optical heterodyne detector that uses local light instead of differential detection using an optical delay circuit, the center wavelength is determined by calculating the amount of wavelength deviation from the received signals of multiple channels. As an optical delay detection circuit to be controlled, for example, an optical heterodyne detector shown in FIG. 33 has been reported (for example, see Non-Patent Document 7).

  In the circuit shown in FIG. 33, two of the four outputs of the optical 90-degree hybrid 2529 are connected to the photoelectric conversion element pair 2504a, and the output of the photoelectric conversion element pair 2504a is Q-arm (quadture- phase-arm (four-phase arm)), and Q-arm includes a low-pass filter (LPF) 2520a. The output of the low-pass filter 2520a is divided into two, and one of the two branched outputs is connected to a discriminator (Dec) 2513a, and the output of the discriminator 2513a is divided into two. One of the branched and bifurcated outputs is connected to the EXOR gate 2514a, and the other of the bifurcated outputs is output as Data1.

  Of the two branched outputs of the low-pass filter 2520a, the other output is branched into two, and one of the two branched outputs is connected to the adder 2522. The output of the device 2522 is connected to an EXOR gate 2514b. The output of the EXOR gate 2514b is connected to the EXOR gate 2514c, and the output of the EXOR gate 2514c is connected to the loop filter 2527. The output of the loop filter 2527 is connected to the local light source 2528, and the output of the local light source 2528 is connected to the optical 90-degree hybrid 2529.

  The other one of the two branched outputs of the low-pass filter 2520a is further divided into two, and the other one of the two branched outputs is connected to the subtractor 2526. The output of the subtracter 2526 is connected to the EXOR gate 2514b.

  On the other hand, the remaining two outputs of the four outputs of the optical 90-degree hybrid 2529 are connected to the photoelectric conversion element pair 2504b, and the output of the photoelectric conversion element pair 2504b is I-arm (In-phase arm (input phase arm). )), And I-arm includes a low-pass filter (LPF) 2520b.

  The output of the low-pass filter 2520b is branched into two, and one of the two branched outputs is connected to a discriminator (Dec) 2513b, and the output of the discriminator 2513b is an EXOR gate 2514a. The output of the EXOR gate 2514a branches into two, one of the two branched outputs is output as Data2, and the other one of the two branched outputs is: It is connected to the EXOR gate 2514c.

  Further, the output of the low-pass filter 2520b is branched into two, and the other one of the two branched outputs is further branched into two. Each branch is added to the adder 2522. And a subtractor 2526.

Housung Yoon et al., "Receiver Structure of Generalized M-ary Optical DPSK System And It's Semi-analytical Performance Evaluation, Tech. 466-467, 2004 C. R. Doerr et al., “Simultaneous reception of both quadratures of 40-Gb / s DQPSK using a simple monochromatic demodulator”, OFC 2005, PDP 12, 2005. Moshe Nazarathy et al., “Stokes Space Optimal Detection of Multiphase Phase and Polarization Shift Keying Modulation”, IEEE J. MoI. Lightwave Technol. , Vol. 24, no. 5, pp. 1978-1988, 2006 Y. Ota et al., “DC-1 Gb / s Burst-Mode Compatible Receiver for Optical Bus Applications”, IEEE J. Lightwave Technol. , Vol. 10, no. 2, pp. 244-249, 1992 Y. Yamada et al., “An Optical 2.5 Gbit / s packet receiving technical for power-fluctuated signals”, CLEO / PACIFIC RIM, TuC3 1000, pp. 8, 1995 H. Toba et al., “A 100-Channel Optical FDM Transmission / Distribution at 622 Mb / s over 50 km”, IEEE J. Biol. Lightwave Technol. , Vol. 8, no. 9, pp. 1396-1401, 1990 S. Norimatsu et al., “An 8Gb / s QPSK Optical Hodyne Detection Exploration External-Cavity Laser Diodes”, IEEE EEPROM. Technol. Lett. , Vol. 4, no. 7, pp 765-767, 1992 JP 2003-44646 A Japanese Patent Laid-Open No. 5-122261

  However, the optical delay detection circuit used in the first to fourth prior art examples needs to have an FSR (Free Spectral Range) having the same frequency as the response speed of the received differential signal. As a result, extremely precise center wavelength (frequency) control is required. This is because if the center wavelength control is not sufficiently performed, the signals between the multiplexed channels leak as noise and degrade the signal quality. For example, a DQPSK (Differential Quadrature Phase Shift Keying) system, which is a first prior art example, requires a center wavelength accuracy of about 1% of FSR in order not to deteriorate transmission quality. However, since a laser light source capable of obtaining an equivalent center wavelength accuracy is expensive, a control circuit that always follows the center wavelength is required on the optical delay detector side.

  In particular, in the first and fifth prior art examples, the more signals are multiplexed, the more precisely the center wavelengths of a plurality of optical delay detection circuits must be simultaneously adjusted, and a plurality of control circuits must be prepared. There was a problem.

  In view of this problem, an optical delay detection circuit using a 2-input 4-output MMI waveguide included in the second prior art example has been devised. According to the second prior art example, if the relative relationship between the center wavelengths of the respective ports is prepared as an initial value, a single control circuit can be used to simultaneously synchronize a plurality of channels.

  However, the phase of the light passing through the MMI waveguide greatly depends on the combination of input and output ports, and in particular, an MMI with a large number of ports has a large number of modes generated in the slab portion, so that a large phase difference occurs. Because of the phase difference depending on the input / output ports, the optical circuit as shown in FIG. 27 cannot precisely match the relative relationship of the center wavelengths of the ports. If the relative relationship between the center wavelengths of each port is not sufficient, some detection is possible, but the signals of each channel multiplexed by multi-level phase modulation cannot be sufficiently separated, and transmission quality deteriorates due to mutual interference. There was a problem of doing.

  In order to accurately match the relative values of the center wavelengths of the ports with the initial values, an optical delay detection circuit using a 2-input 4-output star coupler has been proposed in the third prior art example. When a sufficiently wide star coupler is used, the relative relationship between the center wavelengths of the respective ports can be designed with high accuracy by matching the connection positions of the slab waveguide and the input / output waveguide in the star coupler portion. However, since the slab waveguide is used with the incidence of two light beams, the circuit loss is generally large, and the difference in the amount of light entering the photoelectric conversion element pair is generated. there were.

  As a modulation method that can be received by the optical delay detection circuit other than the optical phase modulation signal, a fourth prior art example that differentially encodes the polarization plane of the signal light has been proposed. Since the fourth prior art example also uses the optical interference in the optical detection circuit to detect, the center wavelength control must be performed precisely as in the optical detection circuit of the optical phase modulation signal, and the multi-level modulation signal is When trying to receive, there has been a problem that a plurality of center wavelength controls are required.

  In addition to the fourth prior art example, besides the optical phase modulation signal, a fifth prior art example for differentially encoding the light intensity is known as a modulation method that can be received by the optical delay detection circuit. In the fifth prior art example, since optical interference is not caused in the optical delay detection circuit, control of center wavelength adjustment is not necessary. However, in order to multiplex signals with high signal-to-noise ratio only by intensity modulation, it is necessary to greatly modulate the signal intensity. As a result, there is a problem that the transmission quality is easily deteriorated due to the non-linearity of the transmission path. It was.

  When creating a practical optical delay detection circuit, it is necessary to automatically monitor the deviation between the wavelength of the signal light and the optimum operating wavelength of the detecting circuit and control the optimum operating wavelength of the detecting circuit.

  In the sixth prior art example, the frequency characteristic of the Mach-Zehnder optical interference circuit and the wavelength of the input signal light are synchronized with a very simple control circuit, but there are various restrictions on the use conditions. First, when the delay time difference generated in the optical detection circuit is larger than the time per bit of the modulation signal, there is a problem that the signal light wavelength information cannot be detected by encoding the modulation signal. This problem can be avoided when the modulation speed is sufficiently low. However, problems may occur even when the modulation rate is sufficiently low. When the Mach-Zehnder optical interference circuit is used as a demultiplexer for a WDM signal, the portion where the frequency response of the Mach-Zehnder optical interference circuit is maximized or minimized corresponds to the optimum wavelength of the optical interference circuit. When a deviation occurs, the absolute value of the wavelength deviation amount can be detected, but correct control cannot be performed because the positive or negative of the deviation amount is not known. In order to avoid this problem, it is necessary to set the control target wavelength of the optical interference circuit to a part where the frequency response of the interference circuit is inclined. In this case, the optimum operation cannot be performed as a demultiplexing circuit. There was a problem of deterioration.

  The seventh prior art example was devised to automatically control the optimum operating wavelength as a duplexer. In the seventh prior art example, the frequency characteristic of the Mach-Zehnder optical interference circuit is actively minutely vibrated, mixed with the AC signal used for the fine vibration and the received signal intensity, and the DC component is extracted by the reduced transmission filter. Is used as a wavelength shift signal. Therefore, it is possible to determine whether the amount of deviation is positive or negative at the optimum wavelength as a duplexer, and it is possible to easily create an automatic control circuit. However, in the seventh prior art example, since the center wavelength of the interferometer is minutely oscillated, there is a problem that interference occurs and quality degradation occurs slightly.

  As described above, the sixth or seventh prior art example is known as a technique for matching the wavelength of the Mach-Zehnder optical interference circuit for multiplexing / demultiplexing the WDM signal with the input light. However, in the optical delay detection circuit that receives the optical multilevel phase modulation signal, the delay time difference is equal to the time per bit of the modulation signal, so the sixth or seventh as shown in FIG. 31 or FIG. The circuit according to the prior art example has a problem that it cannot perform effective center wavelength control.

  On the other hand, in the eighth prior art example, the center wavelength is controlled by calculating the amount of wavelength deviation from the received signals of a plurality of channels. Further, the eighth prior art example is effective even when the delay time difference is equal to the time per bit of the modulation signal. However, the eighth prior art example requires a complicated electric circuit with many components. The eighth prior art example often requires an analog electronic circuit that operates at the same level as the symbol rate. Furthermore, in the eighth prior art example, it is necessary to precisely design and adjust the time constant of each element so that the control loop becomes stable. Therefore, as described above, the eighth prior art example has a problem that there are various factors that make the control circuit expensive.

  The present invention has been made paying attention to the above problems, and an object thereof is to provide an optical delay detection circuit used in an optical multilevel modulation system at low cost.

  In order to achieve the above object, according to a first aspect of the present invention, there is provided an optical delay detection circuit for differentially receiving an input optical signal. A first optical delay unit having a delay unit provides a delay time difference, branches each into N sets (N is 2 or more), inputs to the photoelectric conversion element pair, and at least two sets of the N sets of optical signals. The optical signal is input to the photoelectric conversion element pair after being interfered by the optical multiplexing / demultiplexing circuit after being given different delay time differences by the second optical delay unit.

  Since the optical delay detection circuit as described above uses an optical multiplexing / demultiplexing circuit instead of a slab waveguide, it is possible to reduce the circuit loss and the light quantity balance, and the number of ports of the optical multiplexing / demultiplexing circuit is small. Port dependency is also kept small. By adjusting the delay time difference given by the second optical delay unit, it is possible to design the relative relationship of the center wavelengths with high accuracy for each set of optical signals, and it is generally known such as heater trimming and laser light irradiation. It becomes possible to accurately adjust the relative relationship between the center wavelengths even after the optical waveguide circuit is manufactured by the technology. As a result, if the relative relationship between the center wavelengths can be created or adjusted as an initial value, it is possible to perform batch synchronization of multiple channels by controlling only the delay time difference fine adjustment function unit installed in the first optical delay unit. Only one system is sufficient, and an inexpensive optical delay detection circuit can be provided. Further, by increasing the number of branches N and giving an appropriate optical delay amount by the second optical delay unit, it is possible to receive a transmission signal having a higher multiplicity, such as D8PSK and D16PSK, with a single control circuit. . As a result, the cost per received information amount can be reduced.

  In the aspect corresponding to claim 2 of the present invention, in addition to the aspect corresponding to claim 1 described above, one or more sets of branched optical signals are directly input to the photoelectric conversion element pair. Features.

  In the optical delay detection circuit improved as described above, it is possible to receive not only the optical phase modulation signal but also the intensity-modulated differential optical signal, and more signals can be received by one control circuit. Can be received, and the cost per received information amount can be reduced.

  In the aspect corresponding to claim 3 of the present invention, in addition to the aspect corresponding to either claim 1 or claim 2 of the present invention, the first of the optical signals passing through the second optical delay section is provided. The difference (relative delay time difference) between the delay time difference of the second set of optical signals and the delay time difference of the second set of optical signals is 0.24λ / c to 0.26λ / c (where λ is the wavelength of the optical signal, and c is If the optical speed is set, the DQPSK signal, which is one of the optical multilevel phase modulation signals, can be efficiently received while maintaining the advantage of reducing the cost by simplifying the control circuit.

  Furthermore, in an aspect corresponding to claim 4 of the present invention, polarization rotation is performed before at least one set of optical signals branched into N sets (N is 2 or more) is input to the photoelectric conversion element pair. If the signal passes through the optical path modulation unit or the polarization defining unit, it is possible to receive not only the optical phase modulated signal but also the polarization modulated signal. As a result, more signals can be received by one control circuit, and the cost per received information amount can be reduced.

On the other hand, in the aspect corresponding to claim 5 of the present invention, in addition to the aspect corresponding to any one of claims 1 to 4 of the present invention, one or more sets of optical signals among the branched optical signals are A delay time difference that is α (−1 <α <0) times the delay time difference given by the optical delay unit is given by the second optical delay unit, and a delay time difference fine adjustment function unit is installed in the second optical delay unit, The delay time difference β provided by the first optical delay unit and the delay time difference γ provided by the second optical delay unit operate in conjunction with β while maintaining the relationship of the following equation.
γ = α × β + σ × λ / c (where λ is the wavelength of the input optical signal and σ is a constant having a value of −1 ≦ σ ≦ 1)

  In the above optical delay detection circuit, the sum of the delay time differences given to at least one set of optical signals is a time difference of 0 to 1 times the delay time difference given by the first optical delay unit. For detecting a normal optical multilevel phase modulation signal, the delay time difference provided by the first optical delay unit is used in accordance with the symbol rate of the differential optical signal, so that signals of the same symbol enter the optical multiplexing / demultiplexing circuit at the same time. Become. Interference between the same symbols will have the same optical frequency equivalent characteristics as a normal MZI filter, while the interference output outside the symbol will be equal if it is averaged in a sufficient amount of time. If the signal intensity is monitored and the time average is taken by an integration circuit or the like, the center wavelength shift of the optical delay detection circuit can be detected.

  Since the ratio of interference signals between the same symbols in all output signals is min (−α, α + 1), the wavelength shift can be detected efficiently if designed near α = −0.5.

  Furthermore, in general, a photoelectric conversion element having a large time constant can be obtained at a relatively low cost with a product having a high conversion efficiency. By using such a photoelectric conversion element, the integration circuit can be simplified and a center wavelength shift can be detected. Therefore, the amount of light branched off can be reduced, and the cost of procuring components can be reduced.

  Also, by separating the center wavelength monitor output and the data detection output, by adjusting the delay time difference of the second optical delay unit by adjusting σ, where to set the center wavelength, what is the control system It became possible to choose freely independently.

  In this way, there is no need to vibrate the center wavelength, no complicated and expensive electric circuit is required, and the center wavelength can be easily set to an optimum wavelength as a detection circuit, resulting in deterioration of transmission quality. Therefore, the control circuit can be simplified, and as a result, an inexpensive and high-performance optical delay detection circuit can be provided.

  According to the optical delay detector of the present invention as described above, the control circuit for controlling the center wavelength can be simplified, so that an inexpensive optical delay detector can be provided.

(First embodiment)
FIG. 1 is a block diagram showing an optical delay detection circuit according to the first embodiment of the present invention.

  The optical delay detection circuit according to the first embodiment includes an optical interference circuit portion fabricated on a planar lightwave circuit (PLC), a photoelectric conversion portion composed of a plurality of photodiodes (PD), and an electron that processes an electric signal. It consists of a circuit part.

  As shown in FIG. 1, in the optical interference circuit portion, delay time difference fine adjustment function units 102a and 102b having a structure in which a thin film heater is loaded on a waveguide are connected to the optical branching circuit 101. Here, the delay time difference fine adjustment function units 102 a and 102 b are collectively referred to as a first optical delay unit 103. The output of the delay time difference fine adjustment function unit 102a is connected to the optical N branch circuit 105a, and the output of the delay time difference fine adjustment function unit 102b is connected to the optical N branch circuit 105b. One of the two outputs of the optical N branch circuit 105a is connected to the 2-input 2-output optical multiplexing / demultiplexing circuit 106a via the delay time difference fine adjustment function unit 102c. On the other hand, the other one of the two outputs of the optical N branch circuit 105a is connected to the 2-input 2-output optical multiplexing / demultiplexing circuit 106b via the delay time difference fine adjustment function unit 102e. One of the two outputs of the optical N branch circuit 105b is connected to the 2-input 2-output optical multiplexing / demultiplexing circuit 106a via the delay time difference fine adjustment function unit 102d. On the other hand, the other one of the two outputs of the optical N branch circuit 105b is connected to the 2-input 2-output optical multiplexing / demultiplexing circuit 106b via the delay time difference fine adjustment function unit 102f. Here, the four waveguides including the delay time difference fine adjustment function units 102c, 102d, 102e, and 102f are referred to as a second optical delay unit 107.

Of the four waveguides included in the second optical delay unit 107, only the waveguide connecting the optical N branch circuit 105a and the 2-input 2-output optical multiplexing / demultiplexing circuit 106b is slightly longer than the other three, The lengths of the three waveguides are designed to be equal to each other. Further, the delay time difference, which is a relative difference between the delay time given by the slightly longer waveguide and the delay time given by the other three waveguides, is the used wavelength λ (1580 m), the light velocity c (2.998 × 10 ^8). m / s) and 0.25 × λ ÷ c = 0.318 fsec.

Specifically, the waveguide length is longer than the other three waveguides by the refractive index n (1.455) of the quartz glass waveguide by 0.138 × 10 ¹⁵ × c ÷ n = 0.272 μm. Designed to. Since the received modulation signal bit rate is 21.5 GHz, the delay time difference to be given by the first optical delay unit 103 is designed to be 46.5 ps.

More specifically, a waveguide connected to the optical N branch circuit 105a out of two waveguides included in the first optical delay unit 103 based on the group refractive index ng (1.482) of the quartz glass waveguide. Is designed to be 46.5 × 10 ¹² × c ÷ ng = 9.409 mm longer than the length of the waveguide connected to the optical N branch circuit 105b.

  Note that the delay time difference fine adjustment function unit 102a, 102b, 102c, 102d, 102e, 102f used in the first embodiment of the present invention has a structure in which a thin film heater is loaded on a waveguide. Therefore, the irreversible fine adjustment for correcting the manufacturing error of the waveguide length and the reversible fine adjustment whose self-constant responds in sub-milliseconds can be used separately.

  In the photoelectric conversion portion, photoelectric conversion element pairs 104a and 104b, which are photoelectric conversion modules in which two photodiodes are connected in series on a chip, are used. The output of the 2-input 2-output optical multiplexing / demultiplexing circuit 106a is connected to the photoelectric conversion element pair 104a, and the output of the 2-input 2-output optical multiplexing / demultiplexing circuit 106b is connected to the photoelectric conversion element pair 104b. Inside each photoelectric conversion module, a bias voltage can be applied to both ends of two PDs connected in series, and the voltage between the two PDs can be taken out as an output signal.

  In the electronic circuit portion, the outputs of the photoelectric conversion element pairs 104a and 104b are connected to a decoder 108 that digitizes the received electrical signals and restores the signals encoded for transmission. The output of the photoelectric conversion element pair 104 a is connected to the Q monitor 109. Here, the Q monitor 109 asynchronously samples a part of the output of the photoelectric conversion element pair 104a to estimate the Q value. The output of the Q monitor 109 is connected to the mixer 123, and the output of the mixer 123 is connected to the integrator 110. Further, the output of the integrator 110 is connected to the adder 122, and the output of the adder 122 is connected to the delay time difference fine adjustment function unit 102 a included in the first optical delay unit 103. One of the two outputs of the low-frequency oscillator 124 is connected to the phase adjustment circuit 121, and the other one is connected to the adder 122. The output of the phase adjustment circuit 121 is connected to the mixer 123.

  In the electronic circuit portion, the low frequency oscillator 124 slightly vibrates the center wavelength of the optical interference circuit. Then, the mixer 123 and the integrator 110 take out a wavelength shift correction signal from the output of the Q monitor 109. Then, the adder 122 adds the output of the low frequency oscillator 124 for minute vibration to the wavelength shift correction signal output from the integrator 110. The electrical signal from the adder 122 is designed to be used for the operation of the delay time difference fine adjustment function unit 102 a in the first optical delay unit 103.

  Thus, in the electronic circuit portion, the Q monitor 109, the low frequency transmitter 124, the phase adjustment circuit 121, the mixer 123, the integrator 110, the adder 122, and the delay time difference fine adjustment function unit 102a in the optical interference circuit portion. And operate as the central wavelength control circuit 130 in cooperation with each other.

  When the operation was confirmed after the optical interference circuit portion was manufactured, the center wavelength of the transmittance observed from the output port connected to the 2-input 2-output optical multiplexing / demultiplexing circuit 106a is connected to the 2-input 2-output optical multiplexing / demultiplexing circuit 106b in the design. Only 5.07 GHz was away from the central wavelength of the transmittance observed from the output port by 5.375 GHz (where the wavelength should be 44.8 pm shortwave side), and only 5.07 GHz away. This center wavelength difference corresponds to the fact that the delay time difference given by the second delay unit was given only 0.236λ / c due to a manufacturing error or the like. Therefore, a large voltage is applied for a short time so that an irreversible delay time difference is generated in the delay time difference fine adjustment function unit 102e having a structure in which a thin film heater is loaded on the waveguide, and the delay time difference is constantly added by 0.075 fs. The delay time difference is finely adjusted so that The center wavelength difference after this adjustment was 5.36 GHz even when no voltage was applied. This is equivalent to the difference in delay time given by the second delay unit being 0.249λ / c. In this way, the optical interference circuit portion is constructed by adjusting the relative relationship between the center wavelengths.

  After the optical interference circuit portion, photoelectric conversion portion, and electronic circuit portion are completed, they are assembled into one module.

  After the module is assembled, the frequency dependence of the BER (bit error rate) is evaluated by allocating the wavelength of the input signal light subjected to differential quadrature phase modulation while the central wavelength control circuit is stopped. The frequency dependence of the BER at that time is shown in FIG. The result shown in FIG. 2 was obtained by performing the BER measurement with the signal light center frequency varied by about 4 GHz. The I-ch (In-phase channel (in-phase channel)) signal and the Q-ch (Quadrature-phase channel (orthogonal channel)) signal have a minimum BER at substantially the same frequency. Therefore, it can be seen from FIG. 2 that the relative relationship between the center frequencies and the relative relationship between the center wavelengths of the optical interference circuit portion are sufficiently produced as designed. FIG. 3 shows a result showing the frequency dependence of BER obtained by operating the central wavelength control circuit and performing the same BER measurement. It can be seen from FIG. 3 that the central wavelength control circuit is operating well because the BER is kept at a low value in the range where the signal light center frequency is shifted by about 4 GHz.

  As described above, it has been shown that by using the first embodiment according to the present invention, the wavelength can be controlled sufficiently accurately even with one central wavelength control circuit. Therefore, by using the first embodiment according to the present invention, it is possible to reduce the number of center wavelength control circuits in an optical delay detection circuit that requires a plurality of center wavelength control circuits in the prior art, and is inexpensive. An optical delay detection circuit can be provided.

(Second Embodiment)
FIG. 4 is a block diagram showing an optical delay detection circuit according to the second embodiment of the present invention.

  Similarly to the optical detection circuit according to the first embodiment, the optical detection circuit according to the second embodiment also includes an optical interference circuit portion manufactured on the PLC, a photoelectric conversion portion including a plurality of PDs, and an electric signal. And an electronic circuit portion for processing.

  The optical interference circuit portion includes an optical branching circuit 401, a first optical delay unit 403 including a delay time difference fine adjustment function unit 402a, 402b having a structure in which a thin film heater is loaded on a waveguide, an optical N branch circuit 405a, 405b, 2-input 2-output optical multiplexing / demultiplexing circuits 406a, 406b, 406c, 406d, optical N branching circuits 405a, 405b, and 2-input 2-output optical multiplexing / demultiplexing circuits 406a, 406b, 406c, 406d are connected. A second optical delay unit 407 made of a waveguide is included.

  In the second optical delay unit 407, a delay time difference of λ / 8c is given to the set of optical signals combined by the 2-input 2-output optical multiplexing / demultiplexing circuit 406a and multiplexed by the 2-input 2-output optical multiplexing / demultiplexing circuit 406b. A delay time difference of 3λ / 8c is given to the set of optical signals to be given, and a delay time difference of 5λ / 8c is given to the set of optical signals to be multiplexed by the 2-input 2-output optical multiplexing / demultiplexing circuit 406c. The lengths of the eight waveguides are designed so as to give a delay time difference of 7λ / 8c to the set of optical signals combined by the circuit 406d.

  For example, in order to give a delay time difference of λ / 8c to a set of optical signals multiplexed by the 2-input 2-output optical multiplexing / demultiplexing circuit 406a, an optical N branch circuit 405a and a 2-input 2-output optical multiplexing / demultiplexing circuit 406a are provided. Compared to the waveguide to be connected, the length of the waveguide connecting the optical N branch circuit 405b and the 2-input 2-output optical multiplexing / demultiplexing circuit 406a is calculated from the use wavelength λ (1580 m) and the refractive index n (1.455) by λ ÷. It is designed to be longer by 8 ÷ n = 0.136 μm. Further, since the modulation signal bit rate to be received is 10.5 GHz, the delay time difference to be given by the first optical delay unit 403 is designed to be 95.2 ps.

  Specifically, from the group refractive index ng (1.482) of the quartz glass waveguide, of the two waveguides included in the first optical delay unit 403, the waveguide connected to the optical N branch circuit 405a. The length is designed to be 100 ps × c ÷ ng = 19.266 mm longer than the length of the waveguide connected to the optical N branch circuit 405b.

  After manufacturing the optical interference circuit portion, the transmittance characteristic evaluation is first performed, and then fine adjustment by irradiation with laser light is performed so as to correct the manufacturing error. That is, by irradiating each of the eight waveguides of the second optical delay unit 407 with UV laser light, the center wavelength of the transmittance observed from the output port connected to the 2-input 2-output optical multiplexing / demultiplexing circuit 406a, and 2 The center wavelength of the transmittance observed from the output port connected to the input 2 output optical multiplexing / demultiplexing circuit 406b, the center wavelength of the transmittance observed from the output port connected to the two input 2 output optical multiplexing / demultiplexing circuit 406c, and the 2 input 2 The delay time difference provided by the second optical delay unit 407 is adjusted so that the center wavelength of the transmittance observed from the output port connected to the output optical multiplexing / demultiplexing circuit 406d is aligned at an interval of 2.625 GHz.

  In the second embodiment, a chip having a low conversion frequency but a high conversion efficiency is selected and used for the photoelectric conversion portion, although the cutoff frequency is lower than that of the photoelectric conversion module used in the first embodiment. On the other hand, the photoelectric conversion module according to the second embodiment is the same as that used in the first embodiment. That is, the photoelectric conversion module is configured such that a bias voltage is applied to both ends of two PDs connected in series on the chip, and the voltage between the two PDs becomes an output signal. Such a photoelectric conversion module is used for the photoelectric conversion element pairs 404a, 404b, 404c, and 404d.

  The electronic circuit part includes a decoder 408, an FEC decoder 411, an estimated BER monitor 412, a phase adjustment circuit 421, an adder 422, and a low frequency oscillator 424. Here, the decoder 408 digitizes the received electrical signal and restores the signal encoded for transmission. Then, the FEC decoder 411 performs forward error correction (FEC) using an error correction code added in advance to data before transmission. Then, the estimated BER monitor 412 calculates the ratio of errors corrected from the input / output of the FEC decoder 411. The low frequency oscillator 424 slightly vibrates the center wavelength of the optical interference circuit. The mixer 423 and the integrator 410 extract a wavelength shift correction signal from the electric signal of the low frequency oscillator 424 that has passed through the phase adjustment circuit 421 and the output of the estimated BER monitor 412. Then, the adder 422 adds the output of the low-frequency oscillator 424 for minute vibration to the wavelength shift correction signal output from the integrator 410.

  Here, the electrical signal from the adder 422 is designed to be used for the operation of the delay time difference fine adjustment function unit 402 a in the first optical delay unit 403. The decoder 408 has a logic circuit as shown in FIG. 5, for example. That is, the decoder 408 includes discriminators 513a, 513b, 513c, and 513d that convert signals from the photoelectric conversion elements into digital signals, an EXOR gate 514, and NOT gates 515a and 515b. Here, the threshold values of the discriminators 513a, 513b, 513c, and 513d are set to 0 V of the signals of the photoelectric conversion element pairs 404a, 404b, 404c, and 404d (after removing the DC component).

  After the optical interference circuit portion, the photoelectric conversion portion, and the electronic circuit portion are completed, first, the optical interference circuit portion and the photoelectric conversion portion are connected. Then, an optical differential 8-phase PSK signal with a modulation speed of 10 GHz is input to the connected circuit, and the output from the photoelectric conversion element pair 404a is observed with a digital sampling oscilloscope. At this time, when the wavelength of the input signal is adjusted so that the eye opening is maximized, an eye pattern as shown in FIG. 6 is observed. 6 that the output from the photoelectric conversion element pair is divided into four values.

  When an optical differential 8-phase PSK signal is input to the optical delay detection circuit according to the second embodiment of the present invention, transmission data of the optical differential 8-phase PSK signal, photoelectric element pair output, discriminator output, decoding FIG. 7 shows the relationship between the instrument outputs. When an optical differential 8-phase PSK signal is used, information of 3 bits per symbol can be transmitted. When the relationship between the logical value of each 3 bits and the phase difference between symbols is determined as shown in the first and second columns of FIG. 7, the output from the photoelectric conversion element pair of each channel is from the third column. As shown in the sixth column. Since the threshold values of the discriminators 413a, 413b, 413c, and 413d are set to 0V, the logical values after discrimination are changed from the seventh column to the tenth column. As a result, the logic circuit shown in FIG. The output of the decoder 408 is as shown in the 11th to 13th columns. Since the 3-bit logical values shown in the first column match the decoded logical values in the 11th to 13th columns, the optical delay detection circuit of the second embodiment is completed as designed. Then, it can be seen that the reception detection is correctly performed when the optical differential 8-phase PSK signal is received.

  By using the optical delay detection circuit according to the second embodiment of the present invention, it becomes possible to receive an optical phase differential octet modulation signal having a multiplicity higher than that of the optical phase differential quaternary modulation signal. It is possible to receive more signals with a single control circuit, and it is possible to reduce the cost per received information amount.

(Third embodiment)
FIG. 8 is a block diagram showing an optical delay detection circuit according to the third embodiment of the present invention.

  Similarly to the optical detection circuit according to the first embodiment, the optical detection circuit according to the third embodiment also includes an optical interference circuit part manufactured on the PLC, a photoelectric conversion part including a plurality of PDs, and an electric signal. And an electronic circuit portion for processing.

  In FIG. 8, an optical path is connected to the optical branching circuit 701 and an optical path length fine adjustment function unit 702a having a structure in which a thin film heater is loaded on a waveguide. Here, the delay time difference fine adjustment function unit 702a and the optical path are collectively referred to as a first optical delay unit 703. The output of the delay time difference fine adjustment function unit 702a is connected to the optical N branch circuit 705a, and the output of the optical path is connected to the optical N branch circuit 705b. The output of the optical N branch circuit 705b is connected to the delay time difference fine adjustment function unit 702b.

  The optical interference circuit portion according to the third embodiment has a function of detecting a differential intensity modulation signal, detecting a DQPSK signal, and detecting a center wavelength shift, which are encoded with a modulation speed of 40 GHz.

  The delay time difference of the first optical delay unit 703 according to the third embodiment is designed to be 1 ÷ 40 GHz = 25 ps in accordance with the modulation speed of 40 GHz. Specifically, of the two waveguides included in the first optical delay unit, the waveguide connected to the optical N branch circuit 705a is connected to the other optical N branch circuit 705b. It is designed to be longer by 1 ÷ 40 GHz ÷ c ÷ ng = 5.057 mm. (C is the speed of light, ng is the group refractive index of quartz glass)

  For detection of the differential intensity modulation signal, one set of the four sets of optical signals branched by the optical N branch circuits 705a and 705b is directly passed through the photoelectric conversion element pair without passing through the 2-input 2-output optical multiplexing / demultiplexing circuit. 704a. In the third embodiment, since the 2-input 2-output optical multiplexing / demultiplexing circuit is not used, only the intensity modulation signal, not the phase modulation signal, is received and detected by the photoelectric conversion element pair 704a. That is, comparing the optical signal intensities before and after being given a delay time difference of 25 ps, whichever is greater determines whether the signal from the photoelectric conversion element pair 704a is positive or negative. For such an intensity modulation method, it is sufficient to detect even the difference in the intensity of the optical signal, it is not necessary to make the weaker signal intensity zero, and the amount of information that can be transmitted, regardless of the difference in phase state or polarization state. Although it is half the modulation speed, it can be easily multiplexed with other phase modulation systems and polarization state modulation systems.

On the other hand, of the four sets of optical signals branched by the optical N branch circuits 705a and 705b, two sets of optical signals are used for detection of the DQPSK signal. That is, of the four waveguides connecting the optical N branch circuits 705a and 705b and the two-input two-output optical multiplexing / demultiplexing circuits 706a and 706b, as in the first embodiment, the optical N branch circuit 705a and the two inputs are connected. Only the waveguide connecting the two-output optical multiplexing / demultiplexing circuit 706b is slightly longer than the other three, and the other three waveguides are designed to have the same length. Note that the delay time difference, which is the relative difference between the delay time given by the slightly longer waveguide and the delay time given by the other three waveguides, is the used wavelength λ (1580 m) and the light velocity c (2.998 × 10 ^8). m / s) and 0.25 × λ ÷ c = 0.318 fsec.

  Of the four sets of branched optical signals, the remaining one set of optical signals is used for detecting the center wavelength shift. This set of optical signals is given a delay time difference of 25 ps by the first optical delay unit 703, and then the first optical delay unit 702 by the second optical delay unit 707 including the delay time difference fine adjustment function unit 702b. Half of the delay time difference given in (12.5 ps) is given in the reverse direction, and λ / 8 ÷ c = 0.66 fs is added to adjust the operating point at the center wavelength control, and the second optical delay It is designed to give a delay time difference of −12.5 ps + 0.66 fs = −12.49934 ps in total.

In the third embodiment, a delay time difference that is α (−1 <α <0) times the delay time difference given by the first optical delay unit to one or more sets of branched optical signals is set to the second optical signal. A delay time difference fine adjustment function unit is provided in the second optical delay unit, and the delay time difference β provided by the first optical delay unit and the delay time difference γ provided by the second optical delay unit are given by the delay unit, Operates in conjunction with β while maintaining the relationship of the following equation.
γ = α × β + σ × λ / c (where λ is the wavelength of the input optical signal and σ is a constant having a value of −1 ≦ σ ≦ 1)

  In the third embodiment, α is −0.49997. Since the generation efficiency of the wavelength shift signal is represented by min (−α, α + 1), it can be seen that the wavelength shift signal can be generated and detected with a value substantially close to the maximum value of 0.5. Further, the delay time difference fine adjustment function unit 702a and the delay time difference fine adjustment function unit 702b have the same cross-sectional structure in which the thin film heater is loaded on the waveguide, and the ratio of the lengths of the thin film heaters is 12.49934 ÷ 25 = 0.49997. It is designed to be approximately 0.5. The delay time difference fine adjustment function unit 702a and the delay time difference fine adjustment function unit 702b are connected in series with a gold wire having a small resistance value. Since the delay time difference fine adjustment function unit 702a and the delay time difference fine adjustment function unit 702b are connected in series, the same current flows during operation, and both have the same cross-sectional structure, so the delay time difference fine adjustment function unit 702a. And the delay time difference fine adjustment function unit 702b generate a delay time difference ratio that is the same as the length ratio of each thin film heater. In the third embodiment, λ / 8 ÷ c = 0.66 fs is added to adjust the operating point at the time of center wavelength control in the delay time difference of the second optical delay unit. However, the same effect can be obtained by adding a DC bias power source to the control voltage of the delay time difference fine adjustment function unit 702b and setting σ = 0.125.

  When the equivalent ratio characteristic evaluation of each output port was performed in the vicinity of the optical frequency of 189.7 THz after manufacturing the optical interference circuit portion, the center wavelength of the transmittance observed from the output port connected to the 2-input 2-output optical multiplexing / demultiplexing circuit 706c Was 10 GHz higher than the center wavelength of the transmittance observed from the output port connected to the 2-input 2-output optical multiplexing / demultiplexing circuit 706b, as designed. However, the central wavelength of the transmittance observed from the output port connected to the 2-input 2-output optical branching filter 706c should be 25 GHz higher by design and 27 GHz higher due to manufacturing errors. The manufacturing error is corrected by irradiating the waveguide connecting the circuit 705b and the 2-input 2-output optical multiplexing / demultiplexing circuit 706c with UV laser light.

  For the photoelectric conversion part, the same modules as those in the first embodiment and the second embodiment are used. That is, the photoelectric conversion module is configured such that a bias voltage is applied to both ends of two PDs connected in series on the chip, and the voltage between the two PDs becomes an output signal. Such a photoelectric conversion module is used for the photoelectric conversion element pairs 704a, 704b, 704c, and 704d.

  In the electronic circuit part, the part relating to the demodulation of the data signal is performed by a separate module, and only the integrator 710 necessary for the center wavelength control of the optical interference circuit is built in the electronic circuit part. The integrator 710 removes random noise of inter-bit interference having no center wavelength shift information by extracting high-frequency components from the electrical signal sent from the photoelectric conversion element pair 704d, and extracts center wavelength information. And a portion for generating a signal for controlling the delay time difference fine adjustment function units 702a and 702b from the history accumulation of the difference information between the center wavelength information and the reference voltage (Vth). In the third embodiment, the reference voltage (Vth) is set to the median value of the output fluctuation of the photoelectric conversion element pair 704d (half of the bias voltage applied to both ends of the PD connected in series in the DC component and zero in the AC component). However, depending on the design and manufacturing results of the optical interference circuit part, photoelectric conversion part, electronic circuit part, etc., the output fluctuation range of the photoelectric conversion element pair 704d (excluding the maximum and minimum values), The reference voltage (Vth) may be changed. In particular, when the center wavelength shift detection port of the optical interference circuit portion is shifted due to a manufacturing error, the UV laser beam irradiation step can be omitted if correction is made by changing the setting of the reference voltage (Vth). Therefore, it is effective to provide an inexpensive optical delay detection circuit.

  After the optical interference circuit portion, the photoelectric conversion portion, and the electronic circuit portion are completed, first, the optical interference circuit portion and the photoelectric conversion portion are connected. Then, the wavelength dependence of the photoelectric conversion element pairs 704a, 704b, 704c, and 704d is evaluated using the wavelength tunable laser light that has not been modulated. FIG. 9 shows the frequency dependence of each photoelectric conversion element pair output. In FIG. 9, DA-ch indicates the output of the photoelectric conversion element pair 704a provided for detecting the differential intensity modulation signal. FIG. 9 shows that DA-ch is constant without frequency dependency. The DQPSK signal is detected using the photoelectric conversion element pair 704b (I-ch) and 704c (Q-ch), and each transmission characteristic is a periodic function that vibrates at a period of 40 GHz. It can be seen that the center wavelength is shifted by about 5 GHz. Furthermore, the output (WL-moni) of the photoelectric conversion element pair 704d provided for detecting the center wavelength deviation is a periodic function that vibrates at a period of 80 GHz that is twice that of I-ch or Q-ch. It can be seen that the center wavelength is shifted by about 25 GHz compared to I-ch.

  Here, the delay time difference fine adjustment function unit 702b, the 2-input 2-output optical signal demultiplexer 706c, the photoelectric conversion element pair 704d, and the integrator 710 operate in cooperation as the center wavelength monitor 720.

  10 to 12 are diagrams for explaining control using the center wavelength monitor. The vertical and horizontal dotted lines in FIG. 10 represent the optimum state at a certain point in time. For example, the wavelength of the transmitter is shifted due to a laser failure or the like (arrow 1001), and the center wavelength of the received optical signal is vertical. When the dotted line changes to the vertical solid line, the output of the photoelectric conversion element pair that monitors the wavelength shift increases from the horizontal dotted line to the horizontal solid line (arrow 1002). As shown in FIG. 11, an increase in the operation amount of the delay time difference fine adjustment function units 702a and 702b in the first optical delay unit and the second optical delay unit is instructed based on the output fluctuation of the photoelectric conversion element pair. (Arrow 1101). As a result, as shown in FIG. 12, the transmission characteristic of the optical interference circuit shifts from the dotted line to the solid line (arrow 12001), and it can be seen that the output of the photoelectric conversion element pair returns to the original value again.

  In the optical delay detection circuit of the third embodiment, the part related to the demodulation of the data signal is performed by a separate module, but the optical intensity differential binary signal received and detected by the optical delay detection circuit of the third embodiment With respect to a transmission signal that is redundantly modulated with the optical phase differential quaternary value, the relationship between the redundant modulation signal and the code of each channel will be briefly described. The optical delay detection circuit according to the present embodiment receives, for example, a transmission signal that is redundantly modulated with a light intensity differential binary and an optical phase differential quaternary as shown in FIG. There are two types of light intensity, and there are four types of light phases: -135 degrees, 45 degrees, -45 degrees, and 135 degrees. FIG. 14 shows the relationship between the intensity and phase of the transmission signal and the code of each channel when a signal as shown in FIG. 13 is input to the optical delay detection circuit. First, the respective light intensities and optical phases shown in FIG. 13 are expressed as in the second and third rows. In the optical delay detection circuit of this embodiment, since delay detection is performed, the relative values with the state of the previous signal are as in the fourth row and the fifth row, and finally the light intensity modulation signal (DA-ch) and phase modulation signals (I-ch and Q-ch) are demodulated as in the sixth to eighth rows.

  In this way, the optical intensity modulation signal encoded by the Manchester code can send 1-bit information by two pulses, and the DQPSK signal using the phase 4 value can send 2-bit information by one pulse. Therefore, the transmission signal that is overlap-modulated with the light intensity differential binary and the optical phase differential quaternary as illustrated in FIG. 13 is transmitted at a modulation speed of 40 GHz, and the optical delay detection circuit according to the third embodiment. If it is received, a signal of 100 Gbps can be transmitted.

  A transmission signal that is overlap-modulated with such a light intensity differential binary value and an optical phase differential quaternary value is received by the optical delay detection circuit of the third embodiment, and control is performed using a center wavelength monitor. The graphs of FIGS. 15 to 17 show the eye patterns of signals output from the photoelectric conversion element pairs 704a, 704b, and 704c at the time. The eye pattern of the photoelectric conversion element pair 704a that receives the light intensity modulation signal has a clear eye opening every 50 ps, and the eye pattern of the photoelectric conversion element pair 704b and 704c that receives the phase modulation signal overlaps the intensity modulation signal. For this reason, the line width between the bar and the mark is wide, but a clear eye opening was observed near the median.

  If the optical delay detection circuit according to the third embodiment of the present invention is used, not only the optical phase modulation signal but also the intensity-modulated differential optical signal can be received, and control of one central wavelength system is possible. More signals can be received by the circuit, and the cost per received information amount can be reduced.

(Fourth embodiment)
FIG. 18 is a block diagram showing an optical delay detection circuit according to the fourth embodiment of the present invention.

  The optical detection circuit according to the fourth embodiment includes an optical interference circuit portion having optical components such as a collimator lens and a half mirror, a photoelectric conversion portion including a plurality of PDs, and an electronic circuit portion for processing an electric signal. .

  The optical interference circuit portion includes an optical 2-branch circuit 1201, a first optical delay unit 1203, an optical N-branch circuit 1205, a second optical delay unit 1207, and a 2-input 2-output optical multiplexing / demultiplexing circuit 1206.

  Here, the output of the optical fiber 1231a for spatial system input is connected to a collimator lens 1225a, and the output of the collimator lens 1225a is connected to an optical branching circuit 1201 having a half mirror. Further, the first optical delay unit 1203 includes a delay time difference fine adjustment function unit 1202 having two total reflection mirrors 1230a and 1230b on a straight path stage that is piezoelectrically controlled. The output of the optical N branch circuit 1205 is connected to the two total reflection mirrors 1230c and 1230d. The output of the total reflection mirror 1230c is connected to the fast axis-22.5 degree inclined half-wave plate 1217, and the output of the total reflection mirror 1230d is connected to the fast axis 22.5 degree inclined half-wave plate 1216. ing. Here, the two total reflection mirrors 1230c, 1230d, the two fast axis 22.5 degree inclined half-wave plates 1216, and the fast axis-22.5 degree inclined half-wave plate 1217 are combined to form the second optical delay unit. Called 1207. The output of the second optical delay unit 1207 is connected to a 2-input 2-output optical multiplexing / demultiplexing circuit 1206 for multiplexing interference having a half mirror. The output of the 2-input 2-output optical multiplexing / demultiplexing circuit 1206 is connected to four collimator lenses 1225b, 1225c, 1225d, 1225e for extracting output light. The output of the collimator lens 1225b is connected to the optical fiber 1231b, the output of the collimator lens 1225c is connected to the optical fiber 1231d, the output of the collimator lens 1225d is connected to the optical fiber 1231e, and the output of the collimator lens 1225e is Are connected to the optical fiber 1231c. The outputs of the optical fibers 1231b and 1231c are connected to the photoelectric conversion element pair 1204a, and the outputs of the optical fibers 1231d and 1231e are connected to the photoelectric conversion element pair 1204b.

  The outputs of the photoelectric conversion element pairs 1204a and 1204b are connected to a decoder 1208 that digitizes the received electrical signals and restores the signals encoded for transmission. The output of the photoelectric conversion element pair 1204a is connected to the Q monitor 1209. Here, the Q monitor 1209 estimates the Q value by asynchronously sampling a part of the output of the photoelectric conversion element pair 1204a. The output of the Q monitor 1209 is connected to the mixer 1223, and the output of the mixer 1223 is connected to the integrator 1210. Further, the output of the integrator 1210 is connected to the adder 1222, and the output of the adder 1222 is connected to the delay time difference fine adjustment function unit 1202 included in the first optical delay unit 1203. One of the two outputs of the low-frequency transmitter 1224 is connected to the phase adjustment circuit 1221, and the other one is connected to the adder 1222. The output of the phase adjustment circuit 1221 is connected to the mixer 1223.

  The low frequency oscillator 1224 minutely vibrates the center wavelength of the optical interference circuit. Then, the mixer 1223 and the integrator 1210 extract a wavelength shift correction signal from the output of the Q monitor 1209. Then, the adder 1222 adds the output of the low frequency oscillator 1224 for minute vibration to the wavelength shift correction signal output from the integrator 1210. The electrical signal from the adder 1222 is designed to be used for the operation of the delay time difference fine adjustment function unit 1202 in the first optical delay unit 1203.

  The delay time difference given by the first optical delay unit 1203 is set to 1 / 21.5 GHz = 46.5 ps in accordance with the modulation speed of 21.5 GHz of the received signal. Further, the second optical delay unit 1207 has a fast axis 22.5 only in two optical paths through which light having a smaller delay time provided by the first optical delay unit 1203 among the four optical paths passes. An inclination half-wave plate 1216 and a fast axis-22.5 degree inclination half-wave plate 1217 are inserted. The direction of the fast axis of the fast axis 22.5 degree inclined half-wave plate 1216 is inclined 22.5 degrees from the horizontal plane, and the fast axis of the fast axis-22.5 degree inclined half-wave plate 1217 is inclined by the fast axis 22.5 degrees. It is installed at an angle of −22.5 degrees with respect to the horizontal plane so that it is inclined opposite to the half-wave plate 1216.

  The photoelectric conversion part and the electronic circuit part have exactly the same specifications as those of the optical delay detection circuit of the first embodiment according to the present invention. It is a delay detection circuit module.

  FIG. 19 is a diagram illustrating an example of a polarization state of a transmission signal subjected to optical polarization differential quaternary modulation to be received and detected by the optical delay detection circuit according to the fourth embodiment. The light intensity is subjected to the same modulation for each bit, and is synchronized so that the light intensity decreases when the optical polarization state changes. Such light intensity modulation is not useful for signal multiplexing, but is effective in improving transmission quality because it can remove a chirped light component generated when symbols are switched. The optical polarization state is represented by the polarization angle of linearly polarized light, and one value is selected from 0 (0 degree), π / 2 (90 degrees), π (180 degrees), and 3π / 2 (270 degrees). Take.

  FIG. 20 to FIG. 20 show how such an optical polarization differential four-value modulated transmission signal is detected when input to the optical delay detection circuit of the fourth embodiment. It is the figure which illustrated the relationship between the polarization state of 21 transmission signals, and the code | symbol of each channel.

  20 shows a portion related to the output of the photoelectric conversion element pair 1204a, and FIG. 21 shows a portion related to the output of the photoelectric conversion element pair 1204b. The second row of FIGS. 20 to 21 represents the deflection angle from the horizontal direction of the optical polarization plane of the optical modulation signal exemplified in FIG. A diagram showing the state of the polarization angle of the optical polarization plane when it enters the 2-input 2-output optical multiplexing / demultiplexing circuit 1206 or the 2-input 2-output optical multiplexing / demultiplexing circuit 1206 through a short path (Short-Path). It is shown in the third column of 20-21. The third row and FIG. 20 of FIG. 20 are obtained by the action of the fast axis 22.5 degree inclined half-wave plate 1216 and the fast axis-22.5 degree inclined half-wave plate 1217 inserted in the respective paths. This is a different value from the 21st third column. On the other hand, the deflection angle of the optical polarization plane when it enters the 2-input 2-output optical multiplexing / demultiplexing circuit 1206 or the 2-input 2-output optical multiplexing / demultiplexing circuit 1206 through a long path (Long-Path) does not change to the value itself. Although there is no difference, the first optical delay unit 1203 gives a large delay time difference, so the value of the second row is entered as it is in the column on the right side of the first column. In the 2-input 2-output optical multiplexing / demultiplexing circuit 1206 having a half mirror, the light in the two polarization states displayed in the third column and the fourth column interferes with multiplexing, and thus an index indicating the degree of interference The difference between the declinations displayed in the third column and the fourth column is displayed in the fifth column. 20 and FIG. 21, when the value in the fifth column is between −90 degrees and +90 degrees, the output from the photoelectric conversion element pair 1204b is positive, and at other times, the output is negative.

  If the optical delay detection circuit according to the fourth embodiment of the present invention is used, it is possible to receive a signal subjected to optical polarization differential quaternary modulation. That is, even in an optical delay detection circuit that receives a signal subjected to optical polarization differential multiplex modulation, it becomes possible to receive more signals with a control circuit with one central wavelength, and as a result, per information amount to be received. The cost can be reduced.

(Fifth embodiment)
FIG. 22 is a block diagram showing an optical delay detection circuit according to the fifth embodiment of the present invention.

  Similarly to the first to third embodiments according to the present invention, the optical detection circuit according to the fifth embodiment also includes an optical interference circuit portion manufactured on the PLC, a photoelectric conversion portion including a plurality of PDs, and an electric signal. As in the first embodiment according to the present invention, the optical DQPSK signal modulated at 21.5 GHz is received and detected.

  The optical interference circuit portion includes an optical branching circuit 1401, a first optical delay unit 1403 including a delay time difference fine adjustment function unit 1402a, 1402b having a structure in which a thin film heater is loaded on a waveguide, an optical N branching circuit 1405a, 1405b, 2-input 2-output optical multiplexing / demultiplexing circuits 1406a, 1406b, optical N-branching circuits 1405a, 1405b, and 2-input 2-output optical multiplexing / demultiplexing circuits 1406a, 1406b are connected to delay time difference fine adjustment function units 1402c, 1402d, 1402e, 1402f. A second optical delay unit 1407 composed of four waveguides connected via is included.

  Here, one of the two outputs of the optical N branch circuit 1405a is connected to the 2-input 2-output optical multiplexing / demultiplexing circuit 1406a via the delay time difference fine adjustment function unit 1402c. The other one of the two outputs of the optical N branch circuit 1405a is connected to the 2-input 2-output optical multiplexing / demultiplexing circuit 1406b via the delay time difference fine adjustment function unit 1402e. One of the two outputs of the optical N branch circuit 1405b is connected to the 2-input 2-output optical multiplexing / demultiplexing circuit 1406a via the delay time difference fine adjustment function unit 1402d. The other one of the two outputs of the optical N branch circuit 1405b is connected to the 2-input 2-output optical multiplexing / demultiplexing circuit 1406b via the delay time difference fine adjustment function unit 1402f.

  As the optical interference circuit portion, a portion obtained by partially correcting the optical interference circuit portion used in the first embodiment was used. That is, since the optical wavelength of the carrier wave on which the optical DQPSK signal is superimposed is different between the first embodiment and the fifth embodiment, the delay time difference provided by the second optical delay unit 1407 is 0.25 × 1550 nm ÷ c The layout design has been modified so that = 1.29 fs.

  In the description of the first to third embodiments, it has already been described that the manufacturing error of the optical interference circuit can be adjusted by operating the delay time difference fine adjustment function unit loaded in the second optical delay unit. It is clear that the relative relationship between the center wavelengths of each port can be readjusted with high accuracy even when the wavelength used is changed later.

  For the photoelectric conversion portion, a module group similar to that described in the description of the first to fourth embodiments was used. That is, the photoelectric conversion module is configured such that a bias voltage is applied to both ends of two PDs connected in series on the chip, and the voltage between the two PDs becomes an output signal. Such a photoelectric conversion module is used for the photoelectric conversion element pairs 1404a and 1404b.

  The electronic circuit part calculates the center wavelength shift amount from the analog output from the decoder 1408 that digitizes the received electrical signal and restores the encoded signal for transmission, and the photoelectric conversion element pairs 1404a and 1404b. And a integrator 1410 that accumulates center wavelength shift amount information and generates a signal for controlling the delay time difference fine adjustment function unit 1402a. FIG. 23 is a diagram showing an example of a logic circuit of the phase synchronization circuit 1432 used in the fifth embodiment according to the present invention. As shown in FIG. 23, the logic circuit according to the fifth embodiment includes an adder 1522, a subtractor 1526, and EXOR gates 1514a, 1514b, and 1514c. This is generally known in wireless technology under the name Costas loop, and the amount of deviation can be detected in the baseband without multiplying the modulation signal. In the eighth prior art example, the same logic circuit as in the fifth embodiment is used.

  Finally, after the optical interference circuit portion, the photoelectric conversion portion, and the electronic circuit portion are completed, they are assembled into one module.

  FIG. 24 shows the results of examining the wavelength dependence of the transmittance of the optical interference circuit portion used in the optical delay detection circuit according to the fifth embodiment of the present invention before final assembly. Wavelength dependence curves of transmittance drawn by each output port are as designed by precise PLC manufacturing technology and waveguide length manufacturing error correction technology of delay time difference fine adjustment function part of thin-film heater mounted on waveguide It can be seen that optical interference circuits arranged at regular intervals are obtained.

  FIG. 25 shows a result of examining how the wavelength arrangement of each output port changes when the delay time difference fine adjustment function unit 1402e is manually operated. FIG. 25 shows the operating power dependence of the representative wavelength of each port, with the wavelength that gives the minimum transmittance at each output port as the representative wavelength of each output port. It can be seen that when the power applied to the delay time difference fine adjustment function unit 1402e is increased, the representative wavelength of each port also increases, and the increase is not dependent on the port.

  Even if the center wavelength of the carrier wave varies, the relative phase relationship between the I-ch (In-phase channel (in-phase channel)) and Q-ch (Quadrature-phase channel (quadrature channel)) of the DQPSK signal does not change. The relative frequency difference between the optimum wavelength of the optical delay detection circuit for detecting I-ch and the optimum wavelength of the optical delay detection circuit for detecting Q-ch does not change.

  Even if the center wavelength of the carrier wave of the DQPSK signal has changed due to aging of the laser, wavelength hopping, etc., as shown in FIG. As long as it is built in, only the delay time difference fine adjustment function unit is operated as shown in FIG. 25, and the optimum wavelength of the optical delay detection circuit is simultaneously adjusted again for both I-ch and Q-ch. Is possible. Therefore, according to the fifth embodiment of the present invention, it is not necessary to prepare a plurality of expensive control circuits such as a phase synchronization circuit, so that an inexpensive optical delay detector can be provided.

  In FIG. 34, the result of having calculated the relationship between Q penalty and a relative delay time difference with a simulator is shown.

  Although the Q value representing transmission quality is a function of a plurality of transmission parameters, a numerical value called Q penalty is generally used to extract the dependence of a specific parameter. The definition of the Q penalty is a difference in the Q value indicating how much the Q value has deteriorated when the parameter is changed with reference to the optimum Q value when the parameter is changed.

  From FIG. 34, in order to suppress the degradation of the Q value to 0.3 dB or less, it is added to the first set of optical signals among a plurality of optical signals passing through the second optical delay unit of the optical delay detection circuit. It can be seen that the difference (relative delay time difference) between the delay time difference and the delay time difference applied to the second set needs to be designed between 0.24 and 0.26λ / c. 34, it is necessary to set the relative delay time difference between 0.245 and 0.255λ / c in order to suppress the degradation of the Q value to 0.1 dB or less.

  As described above, according to the optical delay detector of the present invention, the number of control circuits for central wavelength control can be reduced to one, so that an inexpensive optical delay detector can be provided. . In particular, even when the degree of multiplexing of light modulation is increased, the number of required control circuits remains one, so that the cost per received information amount can be reduced. Furthermore, if the center wavelength monitor according to the present invention is used, the control circuit itself is less restricted, and an inexpensive optical delay detector can be provided by simplifying the electric circuit.

1 is a block diagram illustrating an optical delay detection circuit according to a first embodiment of the present invention. In the optical delay detection circuit of 1st Embodiment which concerns on this invention, it is a figure which shows the result of having evaluated the BER by changing a signal light wavelength in the state which stopped the center wavelength control circuit. In the optical delay detection circuit of 1st Embodiment which concerns on this invention, it is a figure which shows the result of having evaluated the BER by changing a signal light wavelength in the state which operated the center wavelength control circuit. It is a block diagram which shows the optical delay detection circuit of 2nd Embodiment which concerns on this invention. It is a block diagram which shows the demodulation circuit used with the optical delay detection circuit of 2nd Embodiment which concerns on this invention. It is a figure which shows an eye pattern when a received signal is tapped in front of a demodulation circuit when an optical differential 8-phase PSK signal is received by the optical delay detection circuit according to the second embodiment of the present invention. It is a figure which shows the relationship between the transmission data of an optical differential 8-phase PSK signal, a photoelectric element pair output, a discriminator output, and a decoder output. It is a block diagram which shows the optical delay detection circuit of 3rd Embodiment concerning this invention. It is a figure which shows the frequency dependence of each photoelectric conversion element pair output at the time of CW light incidence. It is a figure explaining the control using a center wavelength monitor. It is a figure explaining the control using a center wavelength monitor. It is a figure explaining the control using a center wavelength monitor. It is the figure which showed an example about the intensity | strength and phase of a transmission signal which carried out the dual modulation | alteration by optical intensity differential binary and optical phase differential quaternary. It is the figure which illustrated the relationship between the intensity | strength and phase of the transmission signal which carried out the dual modulation | alteration by the optical intensity differential binary and the optical phase differential 4 value, and the code | symbol of each channel. It is a figure which shows an eye pattern when the transmission signal by which the optical intensity differential binary and the optical phase differential quaternary were overlap-modulated was received with the optical delay detection circuit of 3rd Embodiment which concerns on this invention. It is a figure which shows an eye pattern when the transmission signal by which the optical intensity differential binary and the optical phase differential quaternary were overlap-modulated was received with the optical delay detection circuit of 3rd Embodiment which concerns on this invention. It is a figure which shows an eye pattern when the transmission signal by which the optical intensity differential binary and the optical phase differential quaternary were overlap-modulated was received with the optical delay detection circuit of 3rd Embodiment which concerns on this invention. It is a block diagram which shows the optical delay detection circuit of 4th Embodiment concerning this invention. It is the figure which showed an example about the phase and polarization state of the transmission signal which carried out the overlap modulation by the optical phase differential binary modulation and the optical polarization differential modulation. It is the figure which illustrated the relationship between the code | symbol of each phase of the phase and polarization state of the transmission signal which carried out the overlap modulation by the optical phase differential binary modulation and the optical polarization differential modulation. It is the figure which illustrated the relationship between the code | symbol of each phase of the phase and polarization state of the transmission signal which carried out the overlap modulation by the optical phase differential binary modulation and the optical polarization differential modulation. It is a block diagram which shows the optical delay detection circuit of 5th Embodiment concerning this invention. It is a figure which shows the logic circuit of the phase-locked loop circuit used in 5th Embodiment based on this invention. It is the figure which showed the wavelength dependence of the transmittance | permeability of the optical interference circuit part used for the optical delay detection circuit of 5th Embodiment concerning this invention. It is a figure which shows the wavelength dependence change of the optical interference circuit part of 5th Embodiment which concerns on this invention when a delay time difference fine adjustment function part is operated manually. It is a block diagram which shows the 1st prior art example. It is a block diagram which shows the 2nd prior art example. It is a block diagram which shows the 3rd prior art example. It is a block diagram which shows the 4th prior art example. It is a block diagram which shows the 5th prior art example. It is a block diagram which shows the 6th prior art example. It is a block diagram which shows the 7th prior art example. It is a block diagram which shows the 8th prior art example. It is a figure which shows the result of having calculated the relationship between Q penalty and a relative delay time difference with a simulator.

Explanation of symbols

101, 401, 701, 1201, 1401, 1801a, 1801b, 1901, 2001, 2101a, 2101b, 2201, 2301, 2401 Optical branching circuits 102a, 102b, 102c, 102d, 102e, 102f, 402a, 402b, 702a, 702b, 1202, 1402a, 1402b, 1402c, 1402d, 1402e, 1402f, 1802a, 1802b, 1802c, 1802d, 1902a, 1902b, 2002a, 2002b, 2302a, 2302b, 2402a, 2402b Delay time difference fine adjustment function unit 103, 403, 703, 1203 1403, 1803a, 1803b, 1903, 2003, 2103a, 2103b, 2203, 2303, 2403 First optical delay units 104a, 1 4b, 404a, 404b, 404c, 404d, 704a, 704b, 704c, 704d, 1204a, 1204b, 1404a, 1404b, 1804a, 1804b, 1904a, 1904b, 2004a, 2004b, 2104a, 2104b, 2304, 2404, 2504a, 2504b Conversion element pairs 105a, 105b, 405a, 405b, 705a, 705b, 1405a, 1405b, 1205, 1805, 2105 Optical N branch circuits 106a, 106b, 406a, 406b, 406c, 406d, 706a, 706b, 706c, 1206, 1406a, 1406b, 1806a, 1806b, 2106a, 2106b, 2306, 2406 2-input 2-output optical multiplexing / demultiplexing circuits 107, 407, 707, 1207, 14 7 Second optical delay unit 108, 408, 1208, 1408 Decoder 109, 1209 Q monitor 110, 410, 710, 1210, 1410, 2310, 2410 Integrator 411 FEC decoder 412 Estimated BER monitor 513a, 513b, 513c, 513d , 2513a, 2513b Discriminators 514, 1514a, 1514b, 1514c, 2514a, 2514b, 2514c EXOR gates 515a, 515b NOT gate 130 Center wavelength control circuit 720 Center wavelength monitor 1216 Fast axis 22.5 degree inclined half-wave plate 1217 Fast axis − 22.5 degree inclined half-wave plate 2018 Star coupler type optical multiplexing / demultiplexing circuit 1919 2-input 4-output MMI type optical multiplexing / demultiplexing circuits 2320, 2420, 2520a, 2520b Low-pass filters 121, 421, 1221 2421 Phase adjustment circuit 122,422,1222,1522,2422,2522 Adder 123,423,1223,2423 Mixer 124,424,1224,2424 Low frequency oscillator 1225a, 1225b, 1225c, 1225d, 1225e Collimator lens 1526,2526 Subtraction Unit 2527 loop filter 2528 local light source 2529 optical 90-degree hybrid 1230a, 1230b, 1230c, 1230d total reflection mirrors 1231a, 1231b, 1231c, 1231d, 1231e optical fiber 1432 phase synchronization circuit 2133 fast axis 45 degree inclined half-wave plate 2134 fast axis -45 degree inclined half-wave plate

Claims (5)

In an optical delay detection circuit for differential reception by giving a delay time difference to the input optical signal by a first optical delay unit having a delay time difference fine adjustment function unit and then inputting it to the photoelectric conversion element pair,
The optical signals given the delay time difference are branched into N sets (N is 2 or more), respectively, and before being input to the photoelectric conversion element pair, at least two sets of the N sets of optical signals. An optical delay detection circuit characterized in that an optical signal passes through a second optical delay unit that gives a different delay time difference between each set of optical signals and is multiplexed and interfered by an optical multiplexing / demultiplexing circuit.   2. The optical delay detection circuit according to claim 1, wherein at least one set of the optical signals branched into N sets (N is 2 or more) is directly input to the photoelectric conversion element pair. Delay detection circuit.   3. The optical delay detection circuit according to claim 1, wherein, among the optical signals passing through the second optical delay unit, a delay time difference added to a first set of optical signals and a second set of signals are transmitted. An optical delay detection circuit characterized in that the difference (relative delay time difference) from the delay time difference added to the optical signal is set between 0.24λ / c and 0.26λ / c (where λ is the optical signal) Wavelength, c is the speed of light). In an optical delay detection circuit for differential reception, an input optical signal is branched into two and given a delay time difference by a first optical delay unit having a delay time difference fine adjustment function unit and then input to a photoelectric conversion element pair. ,
The optical signals given the delay time difference are branched into N sets (N is 2 or more), respectively, and before the optical signals are input to the photoelectric conversion element pair, at least one of the N sets of optical signals. An optical delay detection circuit, wherein a signal is passed through a polarization rotation unit or a polarization defining unit before being input to a photoelectric conversion element pair. 5. The optical delay detection circuit according to claim 1, wherein at least one of the optical signals branched into N sets (N is equal to or greater than 2) is input to the photoelectric conversion element pair. In the second optical delay unit that passes, a delay time difference that is α (−1 <α <0) times the delay time difference given by the first optical delay unit is given,
In addition, a delay time difference fine adjustment amount γ is given to the set of optical signals by the delay time difference fine adjustment function unit installed in the second optical delay unit,
The fine adjustment amount γ operates in conjunction with the fine adjustment amount β given by the delay time difference fine adjustment function unit installed in the first optical delay unit while maintaining the relationship of the following equation: Delay detection circuit.
γ = α × β + σ × λ / c (where λ is the wavelength of the input optical signal, σ is a constant having a value of −1 ≦ σ ≦ 1, and c is the speed of light)

Images