JP2009171363A - Optical dqpsk receiver and phase control method thereof - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an optical DQPSK receiver further inexpensive and reduced in size and in power consumption, the receiver performing phase control upon an optical DQPSK signal using a Costas loop, and a phase control method thereof. <P>SOLUTION: Photodetectors 31, 32 are disposed in outputs of an optical interferometer 27 for I-phase signal demodulation and an optical interferometer 28 for Q-phase signal demodulation. An electric signal outputs of the photodetectors 31, 32 is partially extracted and connected to limiting amplifiers 33, 34 and low-pass filters (LPFs) 35, 36. LPFs 37, 38 are connected to outputs of the limiting amplifiers 33, 34. An output of the LPF 35 and an output of the LPF 38 are input to a multiplier 39, and an output of the LPF 36 and an output of the LPF 37 are input to a multiplier 40. Outputs of the multipliers 39, 40 are input through averaging circuits 41, 42 to arithmetic circuits 43, 44, inversion processing 45 is applied thereto along a one-side route, and results thereof are connected to inputs of driving circuits 46, 47 for controlling phase adjusters. The driving circuits 46, 47 perform feedback control upon phase adjusters 29, 30, respectively. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an optical DQPSK receiver and a phase control method thereof, and more particularly to an optical DQPSK receiver that performs phase control by a Costas loop and a phase control method thereof.

  The information and communication network has moved away from the conventional telephone network, and is a full IP network (NGN: Next Generation Network) that combines the convenience and economy of the Internet network and the reliability and stability of the telephone network. ) When talking about information communication technology in NGN, digitalization of information, network IP, ubiquitous, and broadband are the keys. The broadbandization of optical communication has progressed dramatically by wavelength multiplexing, but recently the bit rate of time division multiplexing for multiplexing information on the time axis has been remarkable. In conventional wavelength division multiplexing, binary amplitude shift keying (OOF) using the NRZ (Nonreturn-To-Zero) or RZ (Retusn-To-Zero) format has been the mainstream.

  In recent years, modulation / demodulation techniques such as duobinary, CSRZ (Carrier-Suppressed Return-To Zero), and optical DPSK (Differential Phase Shift Keying) have begun to be applied to optical communications. In particular, RZ-DPSK, which modulates the intensity of optical DPSK with RZ, is superior to conventional OOF in terms of deterioration resistance against signal quality degradation due to nonlinear optical effects generated in the transmission optical fiber, and improved S / N by differential reception. This is a modulation / demodulation technique that can also be effective.

  Recently, DQPSK (Differential Quadrature Phase Shift Keying) modulation / demodulation format, which is advantageous for improving the bit rate, has begun to be studied, and is expected to be a promising candidate for future optical communications. This is a binary code that assigns a digital electrical signal with a DPSK code of 0 or 1 to a phase change of light of 0 or π, whereas a DQPSK code has four phase changes (0, π / 2, π, 3π / It is a quaternary code that assigns (0, 0), (0, 1), (1, 0), (1, 1) of the digital electric signal to 2). Therefore, when the same symbol rate is used, DQPSK can allocate twice as much information as DPSK, and the spectral efficiency is doubled. Therefore, the demand for the speed of the electrical device, the adjustment of the dispersion of the optical fiber, Wave mode dispersion is relaxed.

  A typical optical DQPSK receiver (Non-Patent Document 1) includes two Mach-Zehnder interferometers for demodulating I-phase and Q-phase signals. Each Mach-Zehnder interferometer has an optical delay element τ corresponding to the symbol time in the transmission system, and the phase difference with respect to the optical signal provided by the arm of the interferometer is + π / 4 for the I phase and −π / for the Q phase. 4 is set. The two optical output ports of each interferometer are connected to a photodetector comprising two photodiodes that convert the optical signal into an electrical signal.

  Here, in the optical DQPSK receiver, it is important that the phase difference between the arms of the interferometer is accurately set to + π / 4 and −π / 4. The phase error in the interferometer significantly degrades the signal quality Q value. Therefore, in order to obtain an accurate phase difference, the phase is usually controlled by feedback.

  Numerous studies have been made on phase control in demodulation of DQPSK signals in wireless communication systems. A Costas loop is mentioned as a typical phase control method in a wireless communication system (Non-Patent Document 2). FIG. 1 shows a basic configuration of the Costas loop. When controlling the phase of the I phase, extract a part of the output of the main signal of the I phase (A) and multiply it by the Q phase signal component (B) that is orthogonal to the I phase. Take the correlation. The Q phase signal component (B) to be multiplied is used as a reference signal, and in this case, the reference signal is obtained by shifting the output phase of the transmitter by 90 ° and multiplying it with the main signal. . The correlation result (C) includes information on the magnitude and phase direction of the Q phase component contained in the I phase (whether the phase is advanced or delayed), and is ideally compensated for the extra phase in the I phase signal. If the Q-phase component disappears, the correlation result (C) is zero. The control loop feedback-controls the correlation result to the phase (that is, frequency) of the controlled object (in this case, the transmitter) so that the correlation result becomes zero.

A similar Costas loop has been reported in optical DQPSK (Non-patent Document 2). FIG. 2 (Non-Patent Document 3) shows the configuration. This arrangement takes the part of the I-phase main signal (or Q-phase main signal) (s _A or s _B), the reference signal orthogonal (d _A or d _B) and multiplied (multiplier 1 and the multiplier 2) The result is feedback-controlled to the controlled object (in this case, the phase of the interferometer). In principle, it is considered that the configuration is similar to that in the case of a wireless signal in terms of a Costas loop in which I-phase and Q-phase are correlated and fed back. However, in the configuration of an optical signal, the identification circuit 16 and 17 results are used. This is because the generation of the reference signal can be easily realized with a transmitter that performs frequency control with a PLL for a radio signal, whereas the optical signal has a high frequency of THz, so the combination of a PLL and a transmitter is not a practical means. Due to not.

  In this way, by converting optical signals to electrical signals and performing Costas loops in the symbol rate region, it is possible to essentially avoid Q value degradation due to phase fluctuations, which is a problem in the dither-peak-detection method. .

R. A. Griffin and A. C. Carter, “Optical Differential Quadrature Phase-Shift Key (oDQPSK) for High Capacity Optical Transmission”, OFC2002, 2002, pp367-368 ,. Akihiko Sugiura, Bluetooth Information Communication Basics, Interface, August 2001, pp60-68. Z. Tao, A. Isomura, T. Hoshida, and JC Rasmussen, “Dither-free, Accurate, and Robust Phase Offset Monitor and Control Method for Optical DQPSK Demodulator”, ECOC2007, 2007, 3-5-2, pp75- 76. Japanese Patent Application No. 2007-221365

  However, a conventional Costas loop that generates a reference signal using an identification circuit requires an identification circuit that operates at high speed in synchronization with the symbol rate of the main signal. The identification circuit is composed of D flip-flops singly or in combination, and is usually used as an electrical regeneration means for optical signals. However, the higher the clock rate, the higher the cost, the size, and the consumption. There was a problem that power would increase.

  Since optical DQPSK is a modulation / demodulation method that is effective in improving the bit rate compared with optical DPSK, it is assumed that optical DQPSK is used at a high bit rate, and the symbol rate of the main signal is naturally increased. Therefore, the necessity of an identification circuit that matches the symbol rate for generating a reference signal for phase control in optical DQPSK demodulation is an essential problem in terms of price, size, and power consumption.

  The present invention has been made in view of such a problem, and an object of the present invention is to perform optical DQPSK signal phase control using a Costas loop, which is a cheaper, smaller, and lower power consumption optical DQPSK. It is to provide a receiver and a phase control method thereof.

  In order to achieve such an object, the first aspect of the present invention is an optical DQPSK receiver having a phase adjustment mechanism for demodulating an I phase and a Q phase of an optical DQPSK signal. 2, an optical signal output from the first optical interferometer, a first photodetector for converting the optical signal into an electrical signal, and an optical signal output from the second optical interferometer as an electrical signal. A second photodetector that converts the first photodetector, a first limiting amplifier that receives a signal branched from the output of the first photodetector, and a branch from the output of the second photodetector A second limiting amplifier that receives the signal; a first multiplier that multiplies the signal branched from the output of the first photodetector by the output of the second limiting amplifier; And a signal branched from the output of the second photodetector and the first limiting amplifier A second multiplier that multiplies the force by the force, and a first arithmetic circuit that controls the drive circuit of the phase adjustment mechanism of the first optical interferometer so that the output of the first multiplier becomes zero. And a second arithmetic circuit for controlling the drive circuit of the phase adjustment mechanism of the second optical interferometer so that the output of the second multiplier becomes zero.

  The invention according to claim 2 is the optical DQPSK receiver according to claim 1, wherein the saturation output amplitude of the first and second limiting amplifiers and the output amplitude of the first and second photodetectors. Is a ratio of √2 / 2 or less.

  A third aspect of the present invention is the optical DQPSK receiver according to the first or second aspect, wherein the first and second electronic dispersion compensation (EDC) are used in place of the first and second limiting amplifiers. ) A circuit.

  According to a fourth aspect of the present invention, there is provided the optical DQPSK receiver according to any one of the first to third aspects, wherein the optical DQPSK receiver includes low-pass filters that respectively filter signals input to the first and second multipliers. It is characterized by that.

  A fifth aspect of the present invention is the optical DQPSK receiver according to any one of the first to fourth aspects, wherein each of the first and second photodetectors includes two photodiodes. And

  A sixth aspect of the present invention is the optical DQPSK receiver according to any one of the first to fifth aspects, wherein either one of the first and second arithmetic circuits includes an inverting circuit. And

  The invention according to claim 7 is a phase control method for an optical DQPSK receiver, wherein a phase adjustment step for demodulating an I-phase signal and a Q-phase signal from an optical DQPSK signal using an optical interferometer, and an optical DQPSK An optical detection step for converting the I-phase signal and the Q-phase signal demodulated from the signal into electric signals, and a first reference signal and an electric signal in which the amplitude of the I-phase signal converted into the electric signal is limited to a predetermined range. A reference signal generation step for generating a second reference signal in which the amplitude of the Q-phase signal converted into a predetermined range is limited to a predetermined range; and the I-phase signal converted into an electric signal and the second reference signal are multiplied Determining a first phase error amount, multiplying the Q-phase signal converted into an electrical signal by the first reference signal to determine a second phase error amount; and And an inversion step for inverting either of the second phase error signals. The phase adjustment step adjusts the phase of the I-phase signal based on the first phase error amount, and adjusts the phase of the Q-phase signal based on the second phase error amount. .

  The method according to claim 8 is the phase control method of the optical DQPSK receiver according to claim 7, wherein the reference signal generation step includes a saturation output amplitude and an electric signal of the first and second reference signals. The ratio between the output amplitude of the I-phase signal and the Q-phase signal converted to √2 / 2 or less is characterized in that

  According to the present invention, a limiting amplifier is used in place of the expensive, large, and high-power discriminating circuit required for generating the reference signal in constructing the Costas loop, thereby reducing the cost, size, and power consumption. An optical DQPSK receiver can be realized. Furthermore, no matter what the phase of the optical interferometer is, an infinite loop due to hysteresis does not occur, and the phase of the interferometer can be reliably locked to an optimum value.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
[Embodiment 1]
FIG. 3 is a diagram illustrating a phase control method and configuration of the optical DQPSK receiver according to the first embodiment of the present invention. Photodetectors 31 and 32 composed of two photodiodes are arranged at the outputs of the optical interferometer 27 for demodulating the I-phase signal and the optical interferometer 28 for demodulating the Q-phase signal. A part of the electric signal output of the photodetectors 31 and 32 is taken out and connected to the limiting amplifiers 33 and 34 and the low-pass filters 35 and 36. Low-pass filters 37 and 38 are connected to the outputs of the limiting amplifiers 33 and 34. The output of the low pass filter 35 and the output of the low pass filter 38 are input to the multiplier 39. Further, the output of the low-pass filter 36 and the output of the low-pass filter 37 are input to the multiplier 40. The outputs of the multipliers 39 and 40 are input to the arithmetic circuits 43 and 44 through the averaging circuits 41 and 42, and inversion processing 45 is performed on one side path (Q phase in this case) to control the phase adjuster. Connected to the inputs of the drive circuits 46 and 47, the drive circuits 46 and 47 feedback control the phase adjusters 29 and 30 of the optical interferometers 27 and 28, respectively.

  FIG. 4 shows the operation of the limiting amplifier in the Costas loop of the present invention. The dependence of the I and Q phase signal amplitudes on the optical interferometer is shown. Note that the amplitude on the vertical axis is normalized by 1, and the phase unit on the horizontal axis is π.

  The amplitude of the main signal output (signal amplitude at E or F in Fig. 3) is indicated by V (H, H), V (L, L), V (H, L), V (L, H) in Fig. 4. The trigonometric function phase is expressed as a total of four curves shifted by π / 2, π, and 3π / 2 with respect to 0. Ideally, the phases of the I-phase and Q-phase optical interferometers are locked to + π / 4 and −π / 4. In this case, that is, when the phase is + π / 4 and −π / 4, the signal amplitude of the I phase and the Q phase (the signal amplitude at E or F in FIG. 3) is + √2 / 2 in the high state and − in the low state. √2 / 2.

  Here, for example, when the phase I optical interferometer deviates from + π / 4 and shifts in the positive direction, the amplitude curve on the right side is traced from + π / 4 in FIG. 4, and the high level is + √2 / 2 + α and + √2 / 2−α, the low level becomes −√2 / 2−α and −√2 / 2 + α, and both the high level and the low level are split into two amplitudes. A similar splitting phenomenon occurs even when the shift direction is negative. A similar signal amplitude split is also observed by a phase shift on the Q-phase side (−π / 4). Since the signal amplitude split causes the eye to close, the signal quality Q value is significantly degraded.

  Here, consider applying a limiting amplifier to the amplitude signal. The amplitude of the limiting signal output of the main signal (signal amplitude at G or H in FIG. 3) is V (H, H) _LIM, V (L, L) _LIM, V (H, L) _LIM, V in FIG. The curve is represented by (L, H) _LIM. However, in FIG. 4, the coefficient related to the saturation of the limiting amplifier is set to 0.5. That is, the four curves represented by V (H, H) _LIM, V (L, L) _LIM, V (H, L) _LIM, V (L, H) _LIM are V (H, H), The value of 0.5 or more of the four curves V (L, L), V (H, L), and V (L, H) is saturated at 0.5.

Here, an output (signal amplitude at I or J in FIG. 3) obtained by averaging the result of multiplying the main signal output and the limiting amplifier output is considered. This is a value obtained by subtracting the I component 49 of the reference signal from the I component 48 of the Q phase signal in FIG. That is, the monitor output in this case is in the region where cos ω> 0.5.
[V (H, L) −V (H, H)] − [V (L, H) −V (H, H)]
= sinω−cosω−0
= √2sin (ω−π / 4)
It can be expressed. As a result, the monitor output becomes zero when the phase ω of the interferometer is π / 4, and the monitor output becomes a positive value if it deviates in a direction larger than π / 4, and the monitor output becomes a negative value if it deviates in a direction smaller than π / 4. Indicates. When the phase shifts further and reaches cosω <0.5, the monitor output is
[V (H, L) −V (H, H)] − [V (L, H) −V (H, H)]
= sinω−cosω−0.5 + cosω
= sinω−0.5
Will be shown.

  FIG. 5 shows the phase difference dependence of the monitor output (signal amplitude at I or J in FIG. 3) of the optical interferometer. However, the coefficients related to saturation of the limiting amplifier were calculated for 0.3, 0.5, and 0.8. When the coefficients are 0.3 and 0.5, the monitor output becomes zero at π / 4, and shows a tendency to increase monotonically to the right. In other words, when the coefficient is 0.3 and 0.5, if a phase shift occurs from π / 4, the monitor output has a finite value of + or −, and when the monitor output is zero, the phase is uniquely determined to be π / 4. . The processing (for example, proportionality, differentiation, integration processing) necessary for feedback control is performed in the arithmetic circuit 43 or 44 in FIG. 3, and the phase is locked to + π / 4 by setting a control loop so that the monitoring result becomes zero. be able to.

  In this way, by using the output of the limiting amplifier as a reference signal and correlating with the main signal output, the magnitude and direction of the phase shift from π / 4 (whether the phase has shifted in the negative direction from π / 4, Or, it can be understood that the deviation has occurred in the positive direction), and stable feedback control can be realized.

When the coefficient is 0.8, the phase at which the monitor output result is zero is not π / 4 but is about 0.2π to 0.3π. This means that the phase of π / 4 cannot be detected unless a coefficient related to saturation of the limiting amplifier is properly selected. That is, in the region of sinω <0.8, the monitor output is
[V (H, L) −V (H, H)] − [V (L, H) −V (H, H)]
= sinω−cosω−sinω + cosω
= 0
Thus, the phase is not uniquely determined when the monitor output is zero.

  Returning again to FIG. 4, in order to detect the phase of π / 4 in the monitoring result, the value obtained by subtracting the I component 49 of the reference signal from the I component 48 of the Q phase signal is plus about π / 4. It can be seen that the condition is that the relationship is turned back to minus. Accordingly, it is found that the coefficient relating to the saturation of the limiting amplifier needs to be √2 / 2 or less. That is, the saturation output value (absolute value) of the limiting amplifier (the output amplitude of G ′ or H ′ in FIG. 3) is a part of the main signal output amplitude value (absolute value) (E in FIG. 3). The output amplitude of ', F') should be less than √2 / 2 times. Thus, the saturation output range of the limiting amplifier can be obtained from a simple ratio with the branched main signal amplitude. For design and adjustment on the actual circuit, determine the saturation output by selecting the limiting amplifier, or change the branching ratio to change the amplitude value of the main signal output (outputs of E ′ and F ′ in FIG. 3). The amplitude of the main signal output may be determined by linearly amplifying the branched main signal.

  Even if the multipliers 39 and 40, the averaging circuits 41 and 42, and the arithmetic circuits 43 and 44 are constituted by individual analog circuit components, they are converted into digital signals by an A / D conversion circuit once and are digital circuits by FPGA or DSP. Even if configured, the effect obtained by the present invention is the same.

  The inverting circuit 45 is arranged to maintain a relationship in which the I phase and the Q phase are orthogonal to each other, and may be inserted for either the I phase or the Q phase calculation result. Further, the effect obtained by the present invention is the same whether the inverting circuit 45 is constituted by an analog circuit or a digital circuit.

  In the first embodiment described above, the optical interferometers 27 and 28 are made of quartz optical waveguides, and the phase adjustment mechanisms 29 and 30 are provided with metal thin film heaters on the waveguides to control the amount of heat generated by the heaters by the thermo-optic effect. It is extremely effective to make a phase adjuster that adjusts the phase by doing so. Furthermore, a combination of a wave plate and an optical rotator as shown in Patent Document 1 is effective in order to eliminate the polarization dependence of the interferometer due to the silica-based optical waveguide. It is also effective to use the photodetectors 31 and 32 having two photodiodes as a differential receiver.

  Furthermore, although the two optical interferometers 27 and 28 are arranged in a row, the present invention can be applied to other configurations, for example, the configuration of the optical interferometer disclosed in Patent Document 2.

[Embodiment 2]
FIG. 6 is a diagram showing a configuration of an optical DQPSK receiver according to Embodiment 2 of the present invention. Photodetectors 55 and 56 composed of two photodiodes are arranged at the outputs of the optical interferometer 51 for demodulating the I-phase signal and the optical interferometer 52 for demodulating the Q-phase signal. A part of the electric signal output of the photodetectors 55 and 56 is taken out and connected to EDCs 57 and 58 and low-pass filters 59 and 60. Low-pass filters 61 and 62 are branched and connected to the outputs of the EDCs 57 and 58. The output of the low pass filter 59 and the output of the low pass filter 62 are input to the multiplier 63. Further, the output of the low pass filter 60 and the output of the low pass filter 61 are input to the multiplier 64. The outputs of the multipliers 63 and 64 are input to the arithmetic circuits 67 and 68 through the averaging circuits 65 and 66, and are subjected to inversion processing 69 in one path (Q phase in this case) to control the phase adjuster. Connected to inputs of 70 and 71, the drive circuits 70 and 71 feedback control the phase adjusters 53 and 54 of the optical interferometers 51 and 52, respectively. In this configuration, the limiting amplifiers 33 and 34 in the configuration of the first embodiment are replaced with EDCs 57 and 58.

  The operation of the present invention is based on the same principle as in the first embodiment. That is, since the EDCs 57 and 58 are generally composed of a plurality of limiting amplifiers, a plurality of taps, and a plurality of delay circuits, the output signal waveforms of the EDCs 57 and 58 compress the mark level in the same manner as the limiting amplifier output. The waveform is as follows. Therefore, as described in the first embodiment, an error signal can be generated by taking a part of the EDC output as a reference signal and multiplying it with the main output signal component. Here, the saturation output values of the limiting amplifiers of the EDCs 57 and 58 (the output amplitude of G ′ or H ′ in FIG. 3) and the amplitude values of the main signal outputs partially extracted (the E ′ and F ′ in FIG. 3). The effect is maximized by selecting circuit parameters so that the ratio of (output amplitude) is √2 / 2 or less.

  Since the limiting effect of the EDCs 57 and 58 can be effectively used by using the configuration shown in the second embodiment, it is not necessary to newly add a limiting amplifier, and a highly functional and small optical DQPSK receiver can be configured.

  As in the first embodiment, even if the multipliers 63 and 64, the averaging circuits 65 and 66, and the arithmetic circuits 67 and 68 are configured by individual analog circuit components, they are once converted into digital signals by the A / D conversion circuit, Even if it is constituted by a DSP digital circuit, the same effect can be obtained by the present invention. Further, the inverting circuit 69 is disposed in order to maintain a relationship in which the I phase and the Q phase are orthogonal, and may be inserted for either the I phase or the Q phase calculation result. Further, the effect obtained by the present invention is the same whether the inverting circuit 69 is constituted by an analog circuit or a digital circuit.

  Further, in the above-described second embodiment, the optical interferometers 51 and 52 are made of silica-based optical waveguides, and the phase adjusting mechanisms 53 and 54 are provided with metal thin film heaters on the waveguides, and the amount of heat generated by the heaters is controlled by the thermo-optic effect. It is extremely effective to make a phase adjuster that adjusts the phase by doing so. Furthermore, a combination of a wave plate and an optical rotator as shown in Patent Document 1 is effective in order to eliminate the polarization dependence of the interferometer due to the silica-based optical waveguide. It is also effective to use the photodetectors 55 and 56 having two photodiodes as a differential receiver.

  Furthermore, although the two optical interferometers 51 and 52 are arranged in a row, the present invention can be applied to other configurations, for example, the configuration of the optical interferometer disclosed in Patent Document 2.

  In the embodiment, the application to the optical DQPSK signal is described, but the application to the RZ-DQPSK signal is also effective.

  The optical DQPSK receiver and the phase control method thereof according to the present invention can be used for an optical communication device used in an optical communication network or the like.

It is a basic block diagram of the Costas loop in a radio | wireless communications system. It is a block diagram of the Costas loop in optical DQPSK. FIG. 2 is a diagram illustrating a phase control method and a configuration of the optical DQPSK receiver according to the first embodiment. It is a figure which shows operation | movement of the limiting amplifier in the Costas loop of this invention. It is a figure which shows the phase difference dependence of the optical interferometer of the phase monitor output of this invention. 6 is a diagram illustrating a configuration of an optical DQPSK receiver according to Embodiment 2. FIG.

Explanation of symbols

1 Intermediate frequency 2, 3, 6, 20, 21, 39, 40, 63, 64 Multiplier 4, 5, 14, 15, 18, 19, 35, 36, 37, 38, 59, 60, 61, 62 Low pass Filter 7 Loop filter 8 Transmitter 9, 26, 50 Optical DQPSK signal 10, 29, 53 I-phase demodulator optical interferometer phase adjustment mechanism 11, 30, 54 Q-phase demodulator optical interferometer phase adjustment mechanism 12 , 13, 31, 32, 55, 56 Photodetector 16, 17 Identification circuit 22, 23, 41, 42, 65, 66 Averaging circuit 24, 25, 46, 47, 70, 71 Drive circuit 27, 51 I phase Optical interferometer for signal demodulation
28, 52 Optical interferometers 33, 34 for Q-phase signal demodulation Limiting amplifiers 43, 44, 67, 68 Arithmetic circuits 45, 67 Inverting circuit 48 I-component 49 of Q-phase signal 49 I-components 57, 58 EDC of reference signal

Claims (8)

First and second optical interferometers having phase adjustment mechanisms for demodulating the I-phase and Q-phase of the optical DQPSK signal;
A first photodetector for converting an optical signal output from the first optical interferometer into an electrical signal;
A second photodetector for converting an optical signal output from the second optical interferometer into an electrical signal;
A first limiting amplifier that receives a signal branched from the output of the first photodetector;
A second limiting amplifier that receives a signal branched from the output of the second photodetector;
A first multiplier that multiplies the signal branched from the output of the first photodetector by the output of the second limiting amplifier;
A second multiplier that multiplies the signal branched from the output of the second photodetector by the output of the first limiting amplifier;
A first arithmetic circuit that controls a drive circuit of a phase adjustment mechanism of the first optical interferometer so that an output of the first multiplier becomes zero;
An optical DQPSK receiver comprising: a second arithmetic circuit that controls a drive circuit of a phase adjustment mechanism of the second optical interferometer so that an output of the second multiplier becomes zero .   The ratio between the saturation output amplitude of the first and second limiting amplifiers and the output amplitude of the first and second photodetectors is √2 / 2 or less. Optical DQPSK receiver.   3. The optical DQPSK receiver according to claim 1, wherein first and second electronic dispersion compensation (EDC) circuits are used instead of the first and second limiting amplifiers. 4.   The optical DQPSK receiver according to any one of claims 1 to 3, further comprising a low-pass filter that respectively filters signals input to the first and second multipliers.   The optical DQPSK receiver according to any one of claims 1 to 4, wherein each of the first and second photodetectors includes two photodiodes.   6. The optical DQPSK receiver according to claim 1, wherein one of the first and second arithmetic circuits includes an inverting circuit. A phase adjustment step for demodulating the I-phase signal and the Q-phase signal from the optical DQPSK signal using two optical interferometers;
An optical detection step for converting the I-phase signal and the Q-phase signal demodulated from the optical DQPSK signal into electrical signals, respectively;
A first reference signal in which the amplitude of the I-phase signal converted into the electric signal is limited to a predetermined range and a second reference signal in which the amplitude of the Q-phase signal converted into the electric signal is limited to the predetermined range are generated. A reference signal generation step;
The first phase error signal is generated by multiplying the I-phase signal converted into the electrical signal and the second reference signal, and the Q-phase signal converted into the electrical signal is multiplied by the first reference signal. A phase error detection step for generating a second phase error signal;
An inversion step of inverting either of the first and second phase error signals, wherein the phase adjustment step adjusts the phase of the I-phase signal based on the first phase error amount, and A phase control method for an optical DQPSK receiver, wherein phase adjustment of a Q-phase signal is performed based on a phase error amount of 2.   In the reference signal generation step, a ratio between the saturation output amplitudes of the first and second reference signals and the output amplitudes of the I-phase signal and the Q-phase signal converted into electric signals is set to √2 / 2 or less. 8. The phase control method for an optical DQPSK receiver according to claim 7,

Images